Тенденции сокращения потребления липидов, интенсивности лечения и целевого показателя ЛДЛ-С для взрослых, страдающих диабетом и хроническими заболеваниями почек, не связанными с диализом в Гонконге.

Research in context

Evidence before this study

We searched PubMed using the keywords “diabetes,” “CKD,” “lipid management,” “lipid-lowering therapy,” “statins,” “ezetimibe,” “PCSK9i,” and “LDL-C” for original articles and reviews published up to June 8, 2025. Most studies on lipid management trends have predominantly focused on individuals with either diabetes or CKD, leaving a significant gap in understanding lipid management in individuals with both conditions. Since 2022, landmark updates from the Kidney Disease: Improving Global Outcomes (KDIGO) and American Diabetes Association (ADA) have introduced intensive lipid-lowering strategies aimed at atherosclerotic cardiovascular disease (ASCVD) prevention in individuals with coexisting diabetes and CKD, underscoring significant progress in therapeutic approaches. However, despite these advancements, real-world data on LDL-C management in this uniquely high-risk population remain limited.

Added value of this study

This study reports the trends of prescription patterns of lipid-lowering drugs and LDL-C control in individuals with diabetes and non-dialysis CKD in Hong Kong in 2002–2019. Lipid-lowering drugs increased from 17.6% to 71.5%, driven mainly by increased use of statin monotherapy, with low use of ezetimibe or PCSK9i combination therapy. The implementation of recommended treatment intensity remained suboptimal, with 90% of patients not receiving high-intensity therapy for secondary prevention and 70% without taking moderate-intensity therapy for primary prevention in 2019. Although greater improvements in LDL-C attainment were observed over 18 years, less than 30% of individuals attained risk-based LDL-C target recommended by 2019 European Society of Cardiology Guideline.

Implications of all the available evidence

This study highlights treatment gaps between clinical practice and guidelines on LDL-C management among individuals with diabetes and non-dialysis CKD, which call for greater physician awareness and education, review of systematic factors to drug access and prescribing, and alignment of guideline recommendations between professional organizations to maximize cardiovascular protection in this vulnerable population.

Introduction

Individuals with diabetes or chronic kidney disease (CKD) are at high risk of atherosclerotic cardiovascular disease (ASCVD) which increases further when CKD and diabetes co-occur. 1 The relationship between low-density lipoprotein cholesterol (LDL-C) and ASCVD has been confirmed, with 17% reduction in major vascular events for every 1 mmol/L reduction in LDL-C 2 among individuals with diabetes and CKD. Lipid-lowering remains a cornerstone in the multifactorial management of diabetes and CKD. Over the past two decades, newer lipid-lowering drugs (LLDs) such as ezetimibe and proprotein convertase subtilisin/kexin 9 inhibitors (PCSK9i) have expanded treatment options beyond statins. Clinical evidence supports statins for primary and secondary cardiovascular prevention in non-dialysis CKD. 3 In individuals with CKD, combination therapy with statins and ezetimibe has greater efficacy than statin monotherapy in reducing major adverse cardiovascular events (MACE). 4 PCSK9i, given either alone or in combination with statins, reduced LDL-C by up to 60% 5 with 15% reduction in recurrent cardiovascular events in individuals with prior ASCVD. Subgroup analysis indicated similar benefits of PCKS9i in CKD and diabetes subgroups. 6 , 7

Given the high cardiovascular risk in individuals with CKD and with evolving new evidence, clinical guidelines have been updated to emphasize intensive lipid lowering strategies in individuals at high risk of ASCVD. The latest recommendation by the American Diabetes Association (ADA) and Kidney Disease: Improving Global Outcomes (KIDGO) 8 recommended moderate-intensity statin for primary prevention and high-intensity statin for secondary prevention with addition of ezetimibe or PCSK9i as needed, in individuals with CKD and diabetes. Recommendations on LDL-C treatment targets have also evolved, albeit with variations amongst different professional organizations. The Joint European Society of Cardiology (ESC) and European Atherosclerosis Society (EAS) Guideline 9 defined LDL-C treatment targets based on ASCVD risk. The American Cardiology College (ACC) and American Heart Association (AHA) Guidelines 10 recommended use of different intensity of LLD rather than attainment of LDL-C targets. These differences in guideline recommendations may contribute to variations in lipid management, particularly in high-risk individuals with overlapping cardiovascular, kidney, and metabolic conditions.

Individuals with cardiovascular-kidney-metabolic risk factors or diseases are often managed by providers from different specialties, such as cardiology, nephrology, endocrinology, and family medicine, which further contributes to the variations in implementation of guideline-directed medical therapy. Using data from the Hong Kong territory-wide electronic health record (EHR) system, we have reported a 50–70% decline in cardiovascular events among 0.8 million individuals with diabetes, albeit with little change in incidence in CKD. 11 Since lipid lowering is implicated in progression of CKD and prevention of ASCVD in CKD, analysis of pattern of LLD use in individuals with diabetes and CKD may provide new insights. Besides, there are limited real-world studies exploring LDL-C management in individuals with diabetes and non-dialysis CKD. This study explored trends in LDL-C lowering and treatment gaps among these high-risk individuals in Hong Kong from 2002 to 2019. By focusing on this high-risk group, we aimed to provide relevant insights for clinical practice and to improve outcomes.

Methods

Data source

We used data from the Hong Kong territory-wide EHR system to analyze trends in LLD prescriptions and possible gaps between guidelines and clinical practice in individuals with diabetes and non-dialysis CKD. 12 Hong Kong is a cosmopolitan city in South China with 7.5 million residents who are predominantly Chinese. The healthcare system operates on a dual-track basis, comprising both public and private sectors. The government heavily subsidizes the public system and established the Hospital Authority (HA) in 1990 to manage all hospitals and the majority of clinics in the public sector. Over 90% of patients with diabetes receive public healthcare services for a minimal nominal charge, which covers consultations, prescriptions, investigations, and hospitalizations on a per-visit basis (e.g., general outpatient clinic visits have a nominal fee of HK$50 for eligible persons). Based on dataset from the HA-EHR including death date and cause of death, we identified individuals with diabetes from 2000 to 2019 based on any of the following criteria: diagnostic code for diabetes according to the International Classification of Diseases, Ninth Revision (ICD-9) code 250, or according to the Revised Edition of the International Classification of Primary Care (ICPC), codes T89 or T90; glycated hemoglobin (HbA1c) ≥ 6.5% or fasting plasma glucose ≥ 7.0 mmol/L in any available measurement; records of prescription for glucose-lowering drugs; or long-term (≥28 days) insulin prescription. Women with gestational diabetes were excluded. 12 This study was approved by the Joint Chinese University of Hong Kong (CUHK) and New Territories East Cluster (NTEC) Clinical Research Ethics Committee (CREC Ref. NO.: 2020.032, 2023.648) and reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline ( Supplementary Table S1 ). This study utilized comprehensive, anonymized, and de-identified data from the HA-EHR, and informed consent was waived.

Study participants

We included 964,950 individuals with diabetes from 2000 to 2019. To minimize potential biases stemming from incomplete case records during the initial two years following the establishment of the HA-EHR system, we excluded data from 2000 to 2001, yielding 958,079 individuals with diabetes from 2002 to 2019. We excluded individuals aged below 18 years. CKD was defined in individuals with eGFR < 60 mL/min/1.73 m 2 on at least two occasions separated by ≥ 90 days but within 1 year. We estimated kidney function expressed as estimated glomerular filtration rate (eGFR) calculated from plasma creatinine in the EHR using the CKD epidemiological collaboration (CKD-EPI) equation. 13 We excluded eGFR values measured within the same month of hospitalization episodes due to acute kidney injury (AKI) (ICD-9 code: 584). Individuals without any eGFR measurement for that calendar year were excluded from the analysis. We also excluded 9028 individuals undergoing dialysis or kidney replacement therapy identified by ICD-9 codes ( Supplementary Table S2 ). Finally, 332,973 adults with diabetes and non-dialysis CKD were included in the main analysis ( Supplementary Fig. S1 ).

Utilization of LLD

We extracted dispensing data of LLD from the HA-EHR system in 2002–2019. The database provided longitudinal dispensing data of medications including name, dose, frequency, duration (days), start and end dates in 2002–2019. For LLD analysis, we included the following classes of LDL-C regulating drugs within the HA drug formulary: statins (simvastatin, atorvastatin, rosuvastatin, fluvastatin, pravastatin, lovastatin), ezetimibe, and PCSK9i (alirocumab, evolucumab). Supplementary Table S3 lists the Anatomical Therapeutic Chemical (ATC) codes of corresponding drugs. For combination products such as statin-ezetimibe, each drug was counted separately in the utilization analysis. We further considered the proportion of individuals treated with statin monotherapy, and combination therapies (statins plus ezetimibe or statins plus PCSK9i).

Dosage of statins

The dosage of statins was calculated based on total drug dosage and duration of exposure within each calendar year, expressed by:

Individuals with eGFR < 30 mL/min/1.73 m 2 were considered as having advanced CKD. We defined the proportion of individuals with advanced CKD who received an inappropriate dosage of statin, as defined by time-weighted mean daily dosage of simvastatin >40 mg, atorvastatin >80 mg, rosuvastatin >10 mg daily.

We further categorized LLD by LDL-C lowering intensities: low-intensity (LDL-C reduction < 30%), moderate-intensity (LDL-C reduction 30–49%) or high-intensity (LDL-C reduction ≥ 50%). Statin intensities were categorized by types (simvastatin, atorvastatin, rosuvastatin) and time-weighted mean daily dosage during a calendar year. Combination therapies with ezetimibe and PSCK9i were categorized as moderate or high-intensity based on dosage of statins ( Supplementary Table S4 ).

Persistence with statins

We measured the initiation and persistence of statin based on data from the EHR, which includes detailed prescription dispensing data on medication name, dose, frequency, duration (days), and start and end dates. 14 We measured drug persistence using medication possession ratio (MPR) calculated by the proportion of total prescription duration days in each calendar year. Persistent LLD use was defined as a MPR value ≥ 80%. 15

Statistical analysis

All descriptive statistics were reported as counts (percentages) or mean (standard deviation, SD). The age and sex standardized proportions of individuals receiving statin, and dosage regimen for each calendar year were calculated using the 2008 population with diabetes and non-dialysis CKD as the reference further divided by 5-year age groups. Individuals with diabetes and non-dialysis CKD who were alive at the beginning of each year were included in each calendar year. Since there were individuals with missing LDL-C values during a calendar year ( Table 1 ), we only calculated LDL-C target attainment among individuals with at least one LDL-C measurement in each calendar year. We assumed normal distributions of LDL-C values while the proportion of statin use was modeled using a Poisson distribution.

Joinpoint statistical software (Joinpoint Regression Program version 4.8.0.1 [April 2020], Statistical Methodology and Applications Branch, Surveillance Research Program, National Cancer Institute) was used to estimate the trends over time expressed as average-annual-percentage change (AAPC) for the entire observation period and the annual percentage change (APC) during each linear trend segment.

Attainment of treatment intensity recommended by guidelines

Based on the latest recommendation report by the ADA and KDIGO in 2022, 8 all adults with diabetes and CKD are indicated for treatment with at least moderate-intensity LLD. High-intensity LLD is recommended for individuals with ASCVD. We estimated the proportion of individuals who received the recommended LLD treatment intensities for primary and secondary prevention in each calendar year.

Attainment of LDL-C target recommended by guidelines

All laboratories operated by HA were accredited by the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC). We calculated the age- and sex-standardized mean of LDL-C within each calendar year. This enabled us to estimate the proportion of individuals who met recommended LDL-C targets according to the ESC/EAS guidelines. We defined LDL-C target levels primarily according to the latest 2019 ESC/EAS guidelines, 9 which recommend LDL-C targets of <1.8 mmol/L for primary prevention and <1.4 mmol/L for secondary prevention. In addition, given the continued use of earlier recommendations in local clinical practice, we also assessed LDL-C control using the 2016 ESC/EAS guideline targets (<2.6 mmol/L for primary prevention and <1.8 mmol/L for secondary prevention). 16 Additionally, we assessed the proportion of LDL-C target attainment under different treatment intensities (non-LLD users, and users of low, moderate or high-intensity LLD).

Subgroup analysis

We performed subgroup analyses by sex, age-groups (18–49 years, 50–74 years, and ≥75 years), CKD stages, and ASCVD status. We included an 18–49-year age cutoff for subgroup analyses because the KDIGO lipid management guidelines recommend statins for individuals with CKD aged 50 years or older, independent of risk factors, while recognizing the trend toward earlier ASCVD prevention in younger adults. The CKD stages were categorized by eGFR values (CKD-G3: 30–59 ml/min/1.73 m 2 ; CKD-G4: 15–29 ml/min/1.73 m 2 ; CKD-G5: <15 ml/min/1.73 m 2 ). The presence of ASCVD was defined by discharge diagnoses from the earliest hospitalization records in the EHR during the study period (2002–2029), patients were classified as secondary prevention from their first ASCVD hospitalization onward. We defined ASCVD as ischemic heart disease, cerebrovascular disease, or peripheral arterial disease presumed to be of atherosclerotic origin ( Supplementary Table S2 ) according to the ADA 2023 Standards of Medical Care definitions. 17 All statistical analyses were conducted using R version 4.0.2. A two-tailed p value < 0.05 was considered significant.

Results

Over 18 years, the proportion of individuals with diabetes and non-dialysis CKD in Hong Kong increased from 18.1% in 2002 (n = 32,402) to 25.4% (n = 183,907) in 2019 ( Table 1 ). The mean (SD) age of individuals with diabetes and non-dialysis CKD increased from 73.4 (9.6) years in 2002 to 77.9 (10) years in 2019. The respective proportion of men increased from 41.9% to 49.5%. There was a slight decline in prevalence of CKD stage-G5. Prevalence of ASCVD increased from 31.4% in 2002 to 39.9% in 2010, then declined to 35.3% in 2019. The annual mean LDL-C decreased from 3.1 mmol/L in 2002 to 2.0 mmol/L in 2019. The proportion of individuals with missing LDL-C values declined from 66.3% to 22.3% ( Table 1 ).

Utilization and persistence with LLD

Between 2002 and 2019, the overall age- and sex-standardized proportion of statin users showed a steady upward trend (AAPC 8.1 [7.2, 9.8]) rising from 17.6% to 71.3%. This trend was particularly evident in 2007–2011 (APC 16.0 [3.8, 22.3]) ( Supplementary Table S5 ). During the same period, the proportion of individuals with persistent LLD prescription (MPR ≥ 80%) increased from 40.2% to 87.2% with an AAPC of 3.1 [0.7, 5.7]) ( Supplementary Table S6 ). Ezetimibe and PCSK9i were introduced in 2003–2016 respectively. In 2019, the respective usage was 1.19% and 0.01% with 1.00% using statins plus ezetimibe, and 0.01% using statins plus PCSK9i ( Fig. 1 A). Overall, the utilization of statins, ezetimibe, and PCSK9i showed greater increases for secondary than primary prevention ( Supplementary Table S5 ).

Simvastatin use increased over time (AAPC 11.1 [10.0, 12.7]) and remained the most widely used statin (51.9% in 2019). Since 2008, the proportion of atorvastatin use gradually increased to 18.9% in 2019 (AAPC 8.2 [3.4, 14.2]), followed by rosuvastatin (AAPC 13.1 [5.3, 52.3]) as the third most commonly used statin (3.7% in 2019). Pravastatin, lovastatin, and fluvastatin use declined with virtually no usage by 2019 ( Fig. 1 B, Supplementary Table S7 ). Amongst individuals with advanced CKD (CKD G4 and CKD G5), 72.2% of rosuvastatin users received 10 mg or above. This population accounted for 4.6% amongst all statin users in 2019 ( Supplementary Table S8 ).

Guideline-directed treatment intensity in primary and secondary prevention

The majority of individuals received low- (33.4% in 2019) or moderate-intensity therapy (32.6% in 2019) ( Supplementary Table S9 ). For individuals on low-intensity therapy, the majority received simvastatin monotherapy (<20 mg daily). For moderate-intensity therapy, 50% received atorvastatin monotherapy 10–20 mg daily or simvastatin monotherapy 20–40 mg daily. For high-intensity therapy, 50% received atorvastatin monotherapy ≥40 mg daily, and 30% received rosuvastatin monotherapy ≥20 mg daily ( Supplementary Table S10 ). Few individuals received high intensity statin treatment in combination with ezetimibe (10.7%) or PSCK9i (0.4%).

In 2002–2019, the proportion of individuals receiving appropriate LDL-C treatment intensity increased for primary (AAPC 8.9 [7.1, 10.7]) and secondary prevention (AAPC 22.7 [18.8, 30.1]) ( Fig. 2 A and B ). Using the 2019 EAS/ESC treatment guideline, by 2019, only 27.2% received appropriate moderate-intensity therapy for primary prevention, and 11.2% received high-intensity therapy for secondary prevention.

Guideline-directed attainment of LDL-C targets in primary and secondary prevention

The proportion of individuals achieving LDL-C targets increased from 2002 to 2019 with an AAPC of 8.0 [6.5, 11.8] for primary prevention and an AAPC of 10.8 [6.9, 19.7] for secondary prevention. By 2019, for primary prevention, 78.1% of individuals attained LDL-C < 2.6 mmol/L, and 33% attained LDL-C < 1.8 mmol/L. For secondary prevention, 48.8% attained LDL-C < 1.8 mmol/L, and 21.3% attained LDL-C < 1.4 mmol/L ( Fig. 3 A and B ).

Subgroup analysis by age, sex and CKD stages

There were age-related variations in LLD treatment intensity. The 18–49-year age group had the highest proportion of non-users of any LLD (39%) with most of the users receiving high-intensity statins ( Supplementary Table S9 ). Compared to the other age groups, these young individuals were less likely to attain LDL-C target ( Supplementary Table S11 ). The trends for treatment intensity and LDL-C target attainment were similar between men and women.

Over the 18 years, statin use increased across all CKD stages (AAPC 8.3 [7.5, 10] for CKD-G3, AAPC 8.1 [7.4, 9.1] for CKD-G4, AAPC 7.1 [6.0, 8.8] for CKD-G5). By 2019, more individuals with CKD-G3 (71%) received statin monotherapy than individuals with CKD-G4 (69%) and CKD-G5 (64%) ( Supplementary Table S5 ). The latter two groups were more likely to receive combination therapy with statin plus ezetimibe. More individuals in the CKD-G5 (34%) were not on LLD compared with those with CKD-G3 (27.6%) and CKD-G4 (29.4%). All three groups had similar proportions of attaining LDL-C targets. The use of high-intensity LLD was similar across the three CKD stages with individuals treated with low-intensity LLD having the highest proportion of LDL-C target attainment, followed by moderate-intensity, high-intensity, and non-LLD users ( Supplementary Table S11 ).

Discussion

In this Hong Kong territory-wide cohort with diabetes in 2002–2019, we observed increasing survival rates with stable trends of ASCVD but increasing trend of CKD. Amongst 332,975 individuals with diabetes and non-dialysis CKD (G3–G5), the use of LLD had increased from 18% in 2002 to 71% in 2019. By 2019, the use of combination of statins plus ezetimibe or PCSK9i remained less than 1%. The use of LLD was similar across sex and CKD stages (G3–G5) although 39% of the young age group were non-LLD users. Amongst the LLD-users, the majority received low-intensity statin for primary prevention and moderate-intensity statin for secondary prevention. Despite 80% persistence with statin prescription, less than 30% attained risk-based LDL-C targets according to the latest EAS/ESC guidelines.

Most of previous studies reported LLD use only in individuals with diabetes or CKD who had relatively low utilization rates. LLD utilization among individuals with diabetes was reported at 19.5% in Mainland China (2015–2017), 18 45.2% in the United States (US) (2017–2020), 19 and 16.4% in India (2008–2020). 20 Among individuals with CKD, statin use for primary prevention in the US has plateaued at 40% since 2014, 21 with no increase after the 2013 KDIGO guideline. However, data on LLD use among individuals with both diabetes and CKD are scarce. We observed a steady increase in statin use among this high-risk group from 2002 to 2019, with over 70% receiving statins by 2019. Similarly, a survey conducted in the US reported that 68% of individuals with diabetes and non-dialysis CKD received LLD in 2019–2020. 22 These findings suggest that LLD utilization is much higher in individuals with both diabetes and CKD compared to those with either condition alone. In Hong Kong, apart from medical coverage through HA services and availability of generic LLDs, the high proportion of prescription and persistence with statins might be partly due to the initiation of the protocol-driven Risk-Assessment-and-Management-Programme-for-Diabetes-Mellitus (RAMP-DM) in 2009. 23 This program was extended to all primary care clinics since 2013 to facilitate evidence-based management of multiple risk factors. This quality improvement program was implemented against a background of continuous updating and optimization of ASCVD prevention and management guidelines with strong emphasis on LDL-C lowering. In 2007, the ACC/AHA Guideline lowered the LDL-C target and since 2013, the ACC/AHA Guideline 10 proposed a statin-centric approach in order to simplify treatment strategies aimed at increasing the number of statin users.

Variations in statin use by types were likely related to their inclusion into the HA drug formulary, pricing, prevailing evidence, and the availability of generic versions. Simvastatin, the first statin introduced in Hong Kong in 1990, remained the most prescribed statin. Atorvastatin and rosuvastatin use increased steadily after generic drugs were included in the HA drug formulary in 2007 and 2009, respectively. The availability of generics has led to significant price reductions for these medications. The cost per defined daily dose (DDD) is approximately HKD 0.2 (USD 0.03) for generic simvastatin, rosuvastatin and atorvastatin. The price of branded statins is approximately 10 times that of generics. 24 These observations parallel a nationwide Danish cohort (N = 4,998,580) where patent expiry and lower prices increased statin utilization. 25

The cardioprotective effects of statin and ezetimibe combination therapies remained controversial until the publication of the Study of Heart and Renal Protection (SHARP) in 2010, which confirmed the benefits of combination therapy for reducing MACE in individuals with CKD compared to simvastatin alone. Although recommended by clinical guidelines, less than 1% of individuals with diabetes and CKD received ezetimibe in 2019. Low ezetimibe utilization has also been reported in other regions. In a multi-country prospective cohort of individuals with CKD in Brazil, France, Germany, and the US, ezetimibe use was only 0.3–8.5%. 26 Ezetimibe was introduced in Hong Kong as early as 2003, with generic versions available since 2014. Its low use in Hong Kong was not due to lack of access, as the drug was available to individuals not achieving LDL-C targets with statin monotherapy or intolerant of statins. Additionally, the cost of generic ezetimibe is reasonable, at HKD 4.9 (USD 0.6) per 10 mg DDD. Apart from low awareness of clinical evidence and treatment guidelines, the intrinsically low LDL-C levels in individuals with advanced CKD might contribute to under-prescription of LLD.

The prescription of PCSK9i was low which might reflect the late introduction and its high cost. The first PCSK9i, evolocumab, was introduced in Hong Kong on December 6, 2016. 27 In the HA, PCKS9i could only be prescribed by cardiology or endocrinology specialists for secondary prevention of ASCVD in individuals treated with maximally-tolerated statins and suboptimal LDL-C. Additionally, it is indicated for individuals with familial hypercholesterolemia, which has an estimated prevalence of approximately 2.2% in Hong Kong. 28 PCSK9i is substantially more expensive with the cost of evolocumab at HKD 61 (USD 8) per 10 mg DDD and alirocumab at HKD 31 (USD 4) per 5.4 mg DDD. In a meta-analysis of 23 studies, PCSK9i was concluded as being not cost-effective compared to standard LLD even in high-income countries. 29 In FOURIER 6 and ODYSSEY 7 trials, post-hoc analysis stratified by eGFR was performed but there was no dedicated RCT evaluating efficacy of PCSK9i in CKD. The high price, few experience and lack of evidence in individuals with CKD might explain the low usage of PCSK9i compared to statins in our population.

In our cohort, majority of individuals with diabetes and CKD were on low-to-moderate-intensity simvastatin or atorvastatin monotherapy. Only 6% were on high-dose atorvastatin or rosuvastatin. This treatment gap might arise from concerns about adverse effects caused by high-intensity statins such as myotoxicity or elevation in liver enzymes, although the absolute rates of these severe adverse events were extremely low. 30 , 31 Due to pharmacogenetic variations in statin metabolism, exposure to high-dose statins had been associated with 1–2 fold higher plasma levels in Asians compared to Europeans. 32 In Asians, the recommended dose of rosuvastatin should not exceed 20 mg daily with further reduction in the presence of CKD due its renal clearance. The Chinese guideline for lipid management 33 also highlighted limited experience with ‘atorvastatin 80 mg’ in Chinese population and advised its use with caution. In an analysis of pooled data from short and long-term RCTs of atorvastatin in Asians, adverse events from atorvastatin 40 mg daily were rare with few data on 80 mg daily dose. 34

Statin-induced rhabdomyolysis was more common for renally-cleared statins such as rosuvastatin especially if given in high dosage. 35 Although we hypothesized concerns about renal toxicity of high-dose statin might limit its use in advanced CKD, we found similar use of high-intensity LLD across CKD G3-G5 categories. In fact, 72% of individuals with advanced CKD G4-G5 were on inappropriately high dosage of rosuvastatin that exceeded recommendations (>20 mg daily). In advanced CKD, combination of moderate-intensity statins and ezetimibe or PCSK9i may be a safer option for high-intensity LDL-C lowering.

In contrast to other national surveys, we report a relatively high LDL-C attainment rate in Hong Kong with low-intensity statin. In our study of individuals with diabetes and CKD, using LDL-C targets defined by the 2016 ESC guidance, 78% attained the LDL-C target of <2.6 mmol/L for primary prevention and 48% attained <1.8 mmol/L for secondary prevention with 94% of individuals treated with low-to-moderate intensity statins. In a European survey (2017–2018), 18% of individuals with ASCVD attained LDL-C targets 36 with similar rates from US (22.6%, 2021). 37 This response difference might be due to increased sensitivity to statin in Asians. In a meta-analysis, Asians required a lower dose of statin than Europeans to attain comparable reductions in LDL-C and regression in coronary atherosclerotic plaque (rosuvastatin 14.1 mg versus 40 mg). 38 We also observed that LDL-C attainment rates were highest in the low-intensity group, followed by moderate-, high-intensity, and non-users. This trend might be confounded by the lower baseline LDL-C levels in individuals with advanced CKD who were more likely to be given low-intensity statin. However, if we applied the stricter 2019 ESC LDL-C targets, 9 the LDL-C attainment rates dropped to 33% for primary prevention and 21% for secondary prevention. Given that 89% of these high-risk individuals were not on high-intensity statin or combination, there is considerable room for optimizing treatment to reduce risk of ASCVD. The most recent Hong Kong consensus on ASCVD prevention was published in 2016 39 and based on the 2011 ESC/EAS and 2013 ACC/AHA recommendations calling for revision of our local guidelines to close this care gap. 40

Depending on the context, age and gender disparity in statin prescription had been reported. In Netherlands, 41 women were less likely than men to be prescribed statins in primary care setting. In Denmark, 42 the usage of statin was highest in individuals over the age of 75 years. In our cohort, the overall usage of statin was 70% with similar pattern between men and women in Hong Kong. There was a higher proportion of non-statin users (39%) in the 18–49 age group. Most cardiovascular outcome trials in diabetes recruited individuals over the age of 50 with evidence gap in these young individuals. That said, most guidelines recommended the use of statin in individuals under the age of 40 especially in the presence of risk factors for ASCVD. Several surveys had reported suboptimal LDL-C control due to low persistence with statin treatment amongst young individuals, probably due to low motivation or side effects. 43 The high lifetime risk for cardiovascular-renal complications in individuals with young-onset diabetes is now well recognized. 44 This calls for more studies to understand the implementation barriers at both provider and patient levels.

Strengths and limitations

Whilst there are many large-scale surveys on LLD utilization, few studies have specifically examined LLD use in diabetes and CKD. Our study addressed knowledge gaps regarding the low usage of combination statin and non-statin LLD, low usage of LLD in young individuals and use of inappropriately high intensity statin in individuals with advanced CKD. In this 18-year study, the comprehensiveness of the EHR including dispensing, laboratory, and hospitalization data allowed in-depth analysis of the long-term trends and relationships between changing patterns of LLD use and LDL-C control for both primary and secondary prevention of ASCVD.

Our study has several limitations. The EHR-based databases lacked information on the reasons for statin non-prescription and did not include behavioral data on adherence, therefore, we were unable to assess adherence and statin tolerance. We did not have data on types of diabetes, although over 90% of participants were likely to have type 2 diabetes. 45 The LDL-C target attainment analysis was limited to individuals with at least one LDL-C measurement which may introduce selection bias. This low frequency of measurement was particularly evident during the first five years after EHR establishment when benefits of LLD were being evaluated and regular structured assessment of risk factors as a routine quality improvement program was yet to be fully implemented. 23 Additionally, we lacked data on prescriptions from the private sector (<10%), which might lead to underestimation of LLD utilization. CKD stages were classified based on eGFR alone with 40% of individuals in the EHR not having albuminuria data. Other factors such as patients’ and physicians’ perceptions about use of LLD, therapeutic inertia and physicians’ preference for choices of LLD were not assessed. The evaluation of attainment of LDL-C targets and incident ASCVD events and progression of kidney function are beyond the scope of the paper but will be our future work. While our findings share commonalities with studies from other jurisdictions, our conclusions may not be generalizable to other ethnicities or healthcare settings.

Conclusions

In this 18-year survey of a population-based cohort with diabetes and non-dialysis CKD, by 2019, over 70% were treated with statin with high level of persistence. The universal healthcare coverage, introduction of regular structured risk assessment for quality improvement, standardized drug formulary and adherence to practice guidelines was reflected by similar LLD utilization across sex and CKD (G3-G5) categories albeit with the young-age group having the lowest LLD usage. The majority of individuals were treated with low- or moderate-intensity statin monotherapy with only 30% attaining the ESC recommended risk-based LDL-C target in 2019. The low usage of combination of statin and ezetimibe or PCSK9i offered ample opportunities to further reduce LDL-C. These treatment gaps call for greater physician awareness and education, review of systematic factors to optimize LLD prescribing in line with practice guideline to maximize cardiovascular protection in this vulnerable population.

Contributors

YH, AY and EC contributed to conception of the article, statistical analysis, interpretation of results, drafting, revision and approval of the manuscript. MS, HW, CK, JY, and JNML contributed to interpretation of results, revised the manuscript critically and approved the final version. CK, APSK, RCWM, AOYL, and JCNC contributed to conception of the article, revised the manuscript critically and approved the final version. EC, YH and AY have accessed and verified the raw data. EC is the guarantor of this work and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Data sharing statement

The data underlying the results presented in the study are hosted by the Hong Kong Hospital Authority. Due to local regulation, the data are not available to the public. Request for data can be made via Hong Kong Hospital Authority: https://www3.ha.org.hk/data .

Declaration of interests

JCNC reported receiving grants (through institutions) and/or honoraria for consultancy or giving lectures from Abbott, Applied Therapeutics, AstraZeneca, Bayer, Boehringer Ingelheim, Eli Lilly, Hua Medicine, Merck, Novo Nordisk, Powder Pharmaceuticals, Roche, Sanofi, Servier, and Zuellig Pharma Commercialization. APSK has received research grants, travel support, and/or speaker honoraria from Abbott, Astra Zeneca, Bayer, Boehringer Ingelheim, Dexcom, Eli-Lilly, Kyowa Kirin, Merck Serono, Merck Sharp & Dohme, Nestle, Novo-Nordisk, Pfizer, Sanofi and Zuellig Pharma. APSK has participated on Advisoryy Boards for Abbott, Astra Zeneca, Bayer, Biomea, Boehringer Ingelheim, Dexcom, Eli-Lilly, Kyowa Kirin, Merck Sharp & Dohme, Novo-Nordisk, Sanofi and Zuellig Pharma and holds stock in Aptorum. EC has received research grants and/or speaker honoraria from Astra Zeneca, Bayer, Merck KGaA, Hua Medicine, Zuellig Pharma. Other authors have no conflicts of interest to disclose.

Acknowledgements

We acknowledge the Hong Kong Hospital Authority for providing the clinical data.

Part of this work has been presented as a poster abstract at the International Diabetes Federation World Diabetes Congress 2025 in Bangkok, from 7th to 10th April 2025.

Footnotes

Supplementary data related to this article can be found at https://doi.org/10.1016/j.lanwpc.2025.101696 .

Contributor Information

Aimin Yang, Email: aiminyang@cuhk.edu.hk.

Elaine Chow, Email: e.chow@cuhk.edu.hk.

Supplementary data

Supplementary Fig. S1 and Tables S1–S11

References

1. Jin Q., Kuen Lam C.L., Fai Wan E.Y. Association of eGFR slope with all-cause mortality, macrovascular and microvascular outcomes in people with type 2 diabetes and early-stage chronic kidney disease. Diabetes Res Clin Pract. 2023;205 doi: 10.1016/j.diabres.2023.110924.

2. Kearney P.M., Blackwell L., Collins R., et al. Efficacy of cholesterol-lowering therapy in 18,686 people with diabetes in 14 randomised trials of statins: a meta-analysis. Lancet. 2008;371(9607):117–125. doi: 10.1016/S0140-6736(08)60104-X.

3. Liao G., Wang X., Li Y., et al. Antidyslipidemia pharmacotherapy in chronic kidney disease: a systematic review and bayesian network meta-analysis. Pharmaceutics. 2022;15(1) doi: 10.3390/pharmaceutics15010006.

4. Lin Y.C., Lai T.S., Wu H.Y., et al. Effects and safety of statin and ezetimibe combination therapy in patients with chronic kidney disease: a systematic review and meta-analysis. Clin Pharmacol Ther. 2020;108(4):833–843. doi: 10.1002/cpt.1859.

5. Dixon D.L., Pamulapati L.G., Bucheit J.D., et al. Recent updates on the use of PCSK9 inhibitors in patients with atherosclerotic cardiovascular disease. Curr Atheroscler Rep. 2019;21(5):16. doi: 10.1007/s11883-019-0778-6.

6. Charytan D.M., Sabatine M.S., Pedersen T.R., et al. Efficacy and safety of evolocumab in chronic kidney disease in the FOURIER trial. J Am Coll Cardiol. 2019;73(23):2961–2970. doi: 10.1016/j.jacc.2019.03.513.

7. Tuñón J., Steg P.G., Bhatt D.L., et al. Effect of alirocumab on major adverse cardiovascular events according to renal function in patients with a recent acute coronary syndrome: prespecified analysis from the ODYSSEY OUTCOMES randomized clinical trial. Eur Heart J. 2020;41(42):4114–4123. doi: 10.1093/eurheartj/ehaa498.

8. de Boer I.H., Khunti K., Sadusky T., et al. Diabetes management in chronic kidney disease: a consensus report by the American diabetes association (ADA) and kidney disease: improving global outcomes (KDIGO) Diabetes Care. 2022;45(12):3075–3090. doi: 10.2337/dci22-0027.

9. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Atherosclerosis. 2019;290:140–205. doi: 10.1016/j.atherosclerosis.2019.08.014.

10. Stone N.J., Robinson J.G., Lichtenstein A.H., et al. 2013 ACC/AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2014;63(25 Pt B):2889–2934. doi: 10.1016/j.jacc.2013.11.002.

11. Wu H., Lau E.S.H., Yang A., et al. Trends in diabetes-related complications in Hong Kong, 2001-2016: a retrospective cohort study. Cardiovasc Diabetol. 2020;19(1):60. doi: 10.1186/s12933-020-01039-y.

12. Wu H., Lau E.S.H., Yang A., et al. Data resource profile: the Hong Kong diabetes surveillance database (HKDSD) Int J Epidemiol. 2022;51(2):e9–e17. doi: 10.1093/ije/dyab252.

13. Levey A.S., Stevens L.A., Schmid C.H., et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–612. doi: 10.7326/0003-4819-150-9-200905050-00006.

14. De Geest S., Zullig L.L., Dunbar-Jacob J., et al. ESPACOMP medication adherence reporting guideline (EMERGE) Ann Intern Med. 2018;169(1):30–35. doi: 10.7326/M18-0543.

15. Yang A., Shi M., Wu H., et al. Clinical outcomes following discontinuation of metformin in patients with type 2 diabetes and advanced chronic kidney disease in Hong Kong: a territory-wide, retrospective cohort and target trial emulation study. eClinicalMedicine. 2024;71 doi: 10.1016/j.eclinm.2024.102568.

16. Catapano A.L., Graham I., De Backer G., et al. 2016 ESC/EAS guidelines for the management of dyslipidaemias. Rev Esp Cardiol (Engl Ed) 2017;70(2):115. doi: 10.1016/j.rec.2017.01.002.

17. ElSayed N.A., Aleppo G., Aroda V.R., et al. 10. Cardiovascular disease and risk management: standards of care in diabetes-2023. Diabetes Care. 2023;46(Supplement_1):S158–S190. doi: 10.2337/dc23-S010.

18. Zhong V.W., Yu D., Zhao L., et al. Achievement of guideline-recommended targets in diabetes care in China : a nationwide cross-sectional study. Ann Intern Med. 2023;176(8):1037–1046. doi: 10.7326/M23-0442.

19. Chobufo M.D., Regner S.R., Zeb I., Lacoste J.L., Virani S.S., Balla S. Burden and predictors of statin use in primary and secondary prevention of atherosclerotic vascular disease in the US: from the National Health and Nutrition Examination Survey 2017-2020. Eur J Prev Cardiol. 2022;29(14):1830–1838. doi: 10.1093/eurjpc/zwac103.

20. Anjana R.M., Unnikrishnan R., Deepa M., et al. Achievement of guideline recommended diabetes treatment targets and health habits in people with self-reported diabetes in India (ICMR-INDIAB-13): a national cross-sectional study. Lancet Diabetes Endocrinol. 2022;10(6):430–441. doi: 10.1016/S2213-8587(22)00072-9.

21. Iyalomhe O.E., Saparamadu A., Alexander G.C. Use of statins for primary prevention among individuals with CKD in the United States: a cross-sectional, time-trend analysis. Am J Kidney Dis. 2025;85(4):421–431.e1. doi: 10.1053/j.ajkd.2024.11.003.

22. Edmonston D., Lydon E., Mulder H., et al. Concordance with screening and treatment guidelines for chronic kidney disease in type 2 diabetes. JAMA Netw Open. 2024;7(6) doi: 10.1001/jamanetworkopen.2024.18808.

23. Chan J.C., Lim L.-L., Luk A.O., et al. From Hong Kong diabetes register to JADE program to RAMP-DM for data-driven actions. Diabetes Care. 2019;42(11):2022–2031. doi: 10.2337/dci19-0003.

24. Health in action community pharmacy, prescription. https://hia.org.hk/pharmacy/en/medications-1/

25. Wallach Kildemoes H., Vass M., Hendriksen C., Andersen M. Statin utilization according to indication and age: a Danish cohort study on changing prescribing and purchasing behaviour. Health Policy. 2012;108(2–3):216–227. doi: 10.1016/j.healthpol.2012.08.008.

26. Calice-Silva V., Muenz D., Wong M.M.Y., et al. International practice patterns of dyslipidemia management in patients with chronic kidney disease under nephrology care: is it time to review guideline recommendations? Lipids Health Dis. 2023;22(1):67. doi: 10.1186/s12944-023-01833-z.

27. The first PCSK9 inhibitor approved in Hong Kong for the lowering of high cholesterol, 2016-12-06. https://en.prnasia.com/releases/apac/The_First_PCSK9_Inhibitor_Approved_in_Hong_Kong_for_the_Lowering_of_High_Cholesterol-165605.shtml

28. Pang J., Chan D.C., Hu M., et al. Comparative aspects of the care of familial hypercholesterolemia in the “Ten Countries Study”. J Clin Lipidol. 2019;13(2):287–300. doi: 10.1016/j.jacl.2019.01.009.

29. Bagepally B.S., Sasidharan A. Incremental net benefit of lipid-lowering therapy with PCSK9 inhibitors: a systematic review and meta-analysis of cost-utility studies. Eur J Clin Pharmacol. 2022;78(3):351–363. doi: 10.1007/s00228-021-03242-6.

30. Silva M.A., Swanson A.C., Gandhi P.J., Tataronis G.R. Statin-related adverse events: a meta-analysis. Clin Ther. 2006;28(1):26–35. doi: 10.1016/j.clinthera.2006.01.005.

31. NICE Evidence Reviews Collection . National Institute for Health and Care Excellence (NICE) Copyright © NICE; London: 2023. Evidence review for statins: efficacy and adverse effects: cardiovascular disease: risk assessment and reduction, including lipid modification: evidence review C.

32. Tomlinson B., Hu M., Lee V.W., et al. ABCG2 polymorphism is associated with the low-density lipoprotein cholesterol response to rosuvastatin. Clin Pharmacol Ther. 2010;87(5):558–562. doi: 10.1038/clpt.2009.232.

33. Li J.J., Zhao S.P., Zhao D., et al. 2023 Chinese guideline for lipid management. Front Pharmacol. 2023;14 doi: 10.3389/fphar.2023.1190934.

34. Chan J.C., Kong A.P., Bao W., Fayyad R., Laskey R. Safety of atorvastatin in Asian patients within clinical trials. Cardiovasc Ther. 2016;34(6):431–440. doi: 10.1111/1755-5922.12214.

35. Mueller A.M., Liakoni E., Schneider C., et al. The risk of muscular events among new users of hydrophilic and lipophilic statins: an observational cohort study. J Gen Intern Med. 2021;36(9):2639–2647. doi: 10.1007/s11606-021-06651-6.

36. Ray K.K., Molemans B., Schoonen W.M., et al. EU-wide cross-sectional observational study of lipid-modifying therapy use in secondary and primary care: the DA VINCI study. Eur J Prev Cardiol. 2021;28(11):1279–1289. doi: 10.1093/eurjpc/zwaa047.

37. Navar A.M., Shah N.P., Shrader P., et al. Achievement of LDL-C <55 mg/dL among US adults: findings from the cvMOBIUS2 registry. Am Heart J. 2025;279:107–117. doi: 10.1016/j.ahj.2024.06.012.

38. Li Y.F., Feng Q.Z., Gao W.Q., Zhang X.J., Huang Y., Chen Y.D. The difference between Asian and Western in the effect of LDL-C lowering therapy on coronary atherosclerotic plaque: a meta-analysis report. BMC Cardiovasc Disord. 2015;15:6. doi: 10.1186/1471-2261-15-6.

39. Cheung B.M., Cheng C.H., Lau C.P., et al. 2016 Consensus statement on prevention of atherosclerotic cardiovascular disease in the Hong Kong population. Hong Kong Med J. 2017;23(2):191–201. doi: 10.12809/hkmj165045.

40. Yan B.P.Y., Lau K.K., Luk A.O.Y., Wong M.C.S. Call to action: bridging gaps in lipid management in Hong Kong. Hong Kong Med J. 2024;30(2):90–93. doi: 10.12809/hkmj245158.

41. Kiss P.A.J., Uijl A., de Boer A.R., et al. Sex differences in the intensity of statin prescriptions at initiation in a primary care setting. Heart. 2024;110(15):981–987. doi: 10.1136/heartjnl-2023-323722.

42. Wallach-Kildemoes H., Stovring H., Holme Hansen E., Howse K., Pétursson H. Statin prescribing according to gender, age and indication: what about the benefit-risk balance? J Eval Clin Pract. 2016;22(2):235–246. doi: 10.1111/jep.12462.

43. Langslet G., Johansen A.K., Bogsrud M.P., et al. Thirty percent of children and young adults with familial hypercholesterolemia treated with statins have adherence issues. Am J Prev Cardiol. 2021;6 doi: 10.1016/j.ajpc.2021.100180.

44. Nanayakkara N., Curtis A.J., Heritier S., et al. Impact of age at type 2 diabetes mellitus diagnosis on mortality and vascular complications: systematic review and meta-analyses. Diabetologia. 2021;64(2):275–287. doi: 10.1007/s00125-020-05319-w.

45. Yang A., Wu H., Lau E.S., et al. Trends in glucose-lowering drug use, glycemic control, and severe hypoglycemia in adults with diabetes in Hong Kong, 2002–2016. Diabetes Care. 2020;43(12):2967–2974. doi: 10.2337/dc20-0260.

Associated Data

Supplementary Materials

Supplementary Fig. S1 and Tables S1–S11