Association between Low-density Lipoprotein Cholesterol and Sudden Cardiac Arrest in People with Diabetes Mellitus

doi:10.21203/rs.3.rs-2483374/v1

Download PDF

Research Article

Association between Low-density Lipoprotein Cholesterol and Sudden Cardiac Arrest in People with Diabetes Mellitus

https://doi.org/10.21203/rs.3.rs-2483374/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 20 Feb, 2023

Read the published version in Cardiovascular Diabetology →

You are reading this latest preprint version

Background

Dyslipidemia measured as low-density lipoprotein (LDL)-cholesterol is an established risk factor of cardiovascular disease, which is more pronounced in diabetes population. Less is known about the association of LDL-cholesterol level and sudden cardiac arrest (SCA) risk in diabetes mellitus patients. This study investigated the association of LDL-cholesterol level and SCA risk in diabetes population.

Methods

This study was based on Korean National Health Insurance Service database. Patients who received general examination from 2009 to 2012 and diagnosed as type 2 diabetes mellitus were analyzed. Primary outcome was defined as SCA event identified with International Classification of Disease code.

Results

A total of 2,602,577 patients were included, with total follow-up duration of 17,851,797 person*year. Mean follow-up duration was 6.86 years, and 26,341 SCA cases were identified. Overall incidence of SCA was highest in the lowest LDL-cholesterol group (< 70mg/dL) and decreased in a linear manner as LDL-cholesterol rises, till 160mg/dL. Adjustment of covariates resulted in U-shape association, with highest risk of SCA in the highest LDL-cholesterol group (≥ 160mg/dL) followed by lowest LDL-cholesterol group (< 70mg/dL). In subgroup analysis, U-shape association between SCA risk and LDL-cholesterol was more pronounced in male, non-obese people, and those who did not use statins.

Conclusions

In diabetes population, the association between SCA and LDL-cholesterol level was U-shaped with highest and lowest LDL-cholesterol group having higher risk of SCA than others. Paradoxically increased SCA risk in lower LDL-cholesterol population should be recognized and extended to clinical preventive measures.

Lipoproteins

LDL

Diabetes mellitus

Sudden cardiac arrest

Sudden cardiac arrest (SCA) is defined as unexpected sudden loss of cardiac contraction that occurs without warning sign[1, 2]. Although majority of SCA occurs in patients with structural heart disease, including coronary artery disease, heart failure and cardiomyopathy, SCA also affects individuals who are apparently healthy without or with few cardiovascular risk factors[3]. Interlinked with cardiovascular mortality, SCA weighs public health burden that leads to substantial socioeconomic cost globally. Although cardiovascular mortality has decreased in past two decades along with improved preventive measures, cardiovascular death still affects approximately 17 million annual deaths around the world, in which one-fourth presents as SCA[2, 4].

Considering the devastating sequalae of SCA and poor efficacy of secondary prevention, risk stratification of SCA and early recognition of high-risk population is crucial. However, due to sudden onset and dynamic nature, recognizing risk factors of SCA is an area of expertise that needs further identification. For decades, structural heart disease such as ischemic heart disease and reduced left ventricular ejection fraction have been the most widely accepted risk factors of SCA in clinical field, which leads to preventive therapy of implantable cardioverter defibrillator implantation[1, 2, 5]. Besides established structural heart disease, we have identified several independent risk factors of SCA from previous studies, which includes hypertension, diabetes mellitus, dyslipidemia, and metabolic syndrome.[6, 7] Dyslipidemia and diabetes are well-established, modifiable risk factors for future cardiovascular event. Co-existence of diabetes with dyslipidemia elevates the risk of cardiovascular disease, which is also known as diabetic dyslipidemia. Therefore, current guidelines recommend lower target of low-density lipoprotein (LDL)-cholesterol levels in diabetes population, targeting LDL-cholesterol reduction to less than 100mg/dL or even as low as 55mg/dl, depending on individual risk stratification[8, 9]. Although diabetes mellitus and dyslipidemia were also identified as independent predictors of SCA from previous studies, there have been conflicting results on association of dyslipidemia with SCA risk in people with diabetes[10–14]. In addition, less is known about quantitative correlation of LDL-cholesterol level with SCA risk, as well as its inter-connection with presence of diabetes mellitus. Therefore, based on nationwide database of Korean National Health Insurance Service (K-NHIS), we sought to investigate the association between LDL-cholesterol and risk of SCA in diabetes population.

Database

The K-NHIS is the single, exclusive medical insurance system administered by Korean government, that mandates subscription for Korean citizens, including virtually entire Korean population. A regular, nationwide health medical examination is provided for subscribers biennially, which includes physical examinations, self-reported questionnaires regarding sociodemographic factors, and laboratory test. Medical history is recorded as diagnostic codes of International Classification of Disease, 10^th revision (ICD-10), and drug prescription history was recorded. This study was approved by Institutional Review Board of Korea University Anam Hospital and official review committee of the K-NHIS. This study conformed to the principle of 2013 Declaration of Helsinki.

Study Population

Patients with diagnostic codes for type 2 diabetes mellitus and those who underwent medical examination during 2009 to 2012 were included in this study. Patients (i) who were younger than 20 years, (ii) who had diagnostic codes for SCA prior to enrollment, and (iii) those with missing data were excluded. Patients were followed up from the day of initial medical examination to December, 2018. There were no follow-up losses except for death and immigrations.

Definitions Of Variables

The primary outcome was defined as the occurrence of SCA during follow-up and both the aborted and non-aborted SCA events were included. SCA was identified with ICD-10 codes; I46.0 (cardiac arrest with successful resuscitation), I46.1 (sudden cardiac arrest), I46.9 (cardiac arrest, cause unspecified), I49.0 (ventricular fibrillation and flutter), R96.0 (instantaneous death), and R96.1 (death occurring less than 24 hours from onset of symptoms). Out-of-hospital cardiac arrest declared at emergency department was defined as SCA event, and events during in-hospital admission were not counted. Regarding the definition of SCA, patients with prior diagnosis of ischemic stroke, hemorrhagic stroke, asphyxia, suffocation, drowning, anaphylaxis, gastrointestinal bleeding, sepsis, major trauma, hit by lightning, electric shock, or burn within six months of diagnosis of SCA were excluded. The claims for SCA that occurred within one year after health screening was not counted as a main outcome due to ICD-10 coding-based detection of main outcome. For example, claim of SCA codes immediately after health screening can be actual SCA event after health screening or just a repeat claim of SCA which happened before health screening. The robustness of our coding strategy for SCA and other medical conditions was validated in prior studies.[6, 7, 15–19]

Serum lipid profiles were measured as LDL-cholesterol (mg/dL), high-density lipoprotein (HDL)-cholesterol (mg/dL), total cholesterol (mg/dL), and triglyceride (mg/dL) with at least eight hours of fasting. LDL-cholesterol level was classified into eight quantiles: LDL-cholesterol < 70 (group I), 70 ≤ LDL-cholesterol < 75 (group II), 75 ≤ LDL-cholesterol < 100 (group III), 100 ≤ LDL-cholesterol < 115 (group IV), 115 ≤ LDL-cholesterol < 130 (group V), 130 ≤ LDL-cholesterol < 145 (group VI), 145 ≤ LDL-cholesterol < 160 (group VII), 160 ≤ LDL-cholesterol (group VIII). The ICD-10 codes for diagnosis and definitions of variables are described in Supplementary Table S1 and S2.

Statistical analysis

The incidence of SCA was calculated as event numbers per 1,000 person-years of follow-up. Categorical variables were described as number and percentage, and continuous variables were described as mean ± standard deviation or quartiles. Student’s t-test, Mann-Whitney U test, Chi-square test, and Fisher’s exact test was used for comparison of variables as indicated. Cox-proportional hazards model was used to calculate hazards ratios and 95% confidence intervals (CI). Adjustment with covariates was done with multivariate model 1 (age and sex) and model 2 (age, sex, income, body mass index [BMI], smoking status, alcohol consumption status, regular exercise, hypertension, fasting blood glucose, duration of diabetes mellitus, use of insulin, use of oral hypoglycemic agent, and use of statin). All tests were two-tailed, and statistical significance was defined as p-values ≤ 0.05. All statistical analyses were performed with SAS version 9.2 (SAS Institute, Cary, NC, USA).

Study population

During 2009 to 2012, a total of 2,746,079 patients who were diagnosed with diabetes mellitus underwent nationwide medical examination (Fig. 1). Exclusion criteria were (i) Patients under 20 years (n = 390), (ii) those with missing data (n = 117,446), (iii) those with prior diagnosis of SCA during screening period (2002 to 2008; n = 934), or (iv) those who died or experienced SCA within one year after medical examination (n = 24,732). Patients were followed until December 2018, and total follow-up duration was 17,851,797 person*year. Mean follow-up period per patient was 6.86 years, and a total of 26,341 SCA events were identified.

Baseline characteristics regarding LDL-cholesterol level is described in Table 1. As LDL-cholesterol level decreases, there was tendency of increase of age, and decrease of BMI. In addition, lower LDL-cholesterol subgroups resulted higher proportion of male sex, current-smoker, heavy-drinker, and patients with regular exercise. Lower LDL-cholesterol subgroup also reflected higher proportion of pre-existing comorbidities of hypertension, and long-standing diabetes mellitus (more than five years), as well as prior medication of statin, multiple oral hypoglycemic agents (three or more), and insulin.

Table 1

Baseline characteristics according to LDL-cholesterol level
	LDL-cholesterol (mg/dL)
	< 70	< 75	< 100	< 115	< 130	< 145	< 160	≥ 160	p-value
	343,781	305,842	389,404	415,948	380,029	297,720	203,636	266,217
Male	235,091 (68.4%)	192,013 (62.7%)	242,681 (62.3%)	256,338 (61.6%)	227,758 (59.9%)	170,905 (57.4%)	110,355 (54.1%)	127,157 (47.7%)
Age (years)	58.0 ± 12.1	58.2 ± 12.4	57.6 ± 12.5	57.3 ± 12.5	57.1 ± 12.4	56.9 ± 12.3	56.9 ± 12.1	56.9 ± 11.9	< 0.001
Age groups									< 0.001
< 40 years	23,448 (6.8%)	22,404 (7.3%)	30,930 (7.9%)	33,586 (8.0%)	29,972 (7.8%)	22,736 (7.6%)	15,137 (7.4%)	18,440 (6.9%)
40–64 years	211,359 (61.4%)	182,765 (59.7%)	236,222 (60.6%)	256,037 (61.5%)	237,927 (62.6%)	189,987 (63.8%)	131,252 (64.4%)	174,977 (65.7%)
≥ 65 years	108,974 (31.7%)	100,673 (32.9%)	122,252 (31.3%)	126,325 (30.3%)	112,130 (29.5%)	84,997 (28.5%)	57,247 (28.1%)	72,800 (27.3%)
BMI (kg/m²)	24.9 ± 3.4	24.8 ± 3.4	24.8 ± 3.4	24.9 ± 3.4	25.1 ± 3.3	25.2 ± 3.3	25.3 ± 3.3	25.4 ± 3.3	< 0.001
BMI groups									< 0.001
< 18.5	7,456 (2.1%)	6,480 (2.1%)	7,557 (1.9%)	6,856 (1.6%)	5,188 (1.3%)	3,519 (1.1%)	2,004 (0.9%)	2,538 (0.9%)
< 23	89,058 (25.9%)	82,116 (26.8%)	103,622 (26.6%)	107,561 (25.8%)	93,542 (24.6%)	69,772 (23.4%)	45,379 (22.2%)	57,156 (21.4%)
< 25	83,318 (24.2%)	75,096 (24.5%)	96,421 (24.7%)	103,432 (24.8%)	95,642 (25.1%)	74,654 (25.0%)	50,990 (25.0%)	65,891 (24.7%)
< 30	139,334 (40.5%)	120,471 (39.3%)	153,721 (39.4%)	167,244 (40.2%)	155,697 (40.9%)	125,535 (42.1%)	87,892 (43.1%)	116,545 (43.7%)
≥ 30	24,615 (7.1%)	21,679 (7.0%)	28,083 (7.2%)	30,855 (7.4%)	29,960 (7.8%)	24,240 (8.1%)	17,371 (8.5%)	24,087 (9.0%)
Waist circumference (cm)	85.8 ± 8.7	85.2 ± 8.8	85.1 ± 8.7	85.2 ± 8.7	85.4 ± 8.6	85.5 ± 8.5	85.6 ± 8.4	85.6 ± 8.5	< 0.001
Income, Lowest Q1	73,235 (21.3%)	63,875 (20.8%)	80,925 (20.7%)	87,028 (20.9%)	79,498 (20.9%)	62,148 (20.8%)	42,551 (20.9%)	57,194 (21.4%)	< 0.001
Smoking									< 0.001
Non-smoker	168,846 (49.1%)	165,626 (54.1%)	213,368 (54.7%)	229,606 (55.2%)	213,254 (56.1%)	170,622 (57.3%)	120,511 (59.1%)	166,465 (62.5%)
Ex-smoker	70,670 (20.5%)	60,372 (19.7%)	74,524 (19.1%)	79,039 (19.0%)	70,311 (18.5%)	52,712 (17.7%)	34,019 (16.7%)	39,383 (14.7%)
Current-smoker	104,265 (30.3%)	79,844 (26.1%)	101,512 (26.0%)	107,303 (25.8%)	96,464 (25.3%)	74,386 (24.9%)	49,106 (24.1%)	60,369 (22.6%)
Drinking									< 0.001
Non-drinker	175,256 (50.9%)	172,941 (56.5%)	219,889 (56.4%)	235,592 (56.6%)	218,788 (57.5%)	175,221 (58.8%)	123,371 (60.5%)	170,436 (64.0%)
Mild-drinker	118,004 (34.3%)	99,584 (32.5%)	129,764 (33.3%)	139,985 (33.6%)	126,755 (33.3%)	97,033 (32.5%)	64,152 (31.5%)	76,820 (28.8%)
Heavy-smoker	50,521 (14.7%)	33,317 (10.8%)	39,751 (10.2%)	40,371 (9.7%)	34,486 (9.0%)	25,466 (8.5%)	16,113 (7.9%)	18,961 (7.1%)
Regular exercise	73,598 (21.4%)	66,558 (21.7%)	83,520 (21.4%)	87,235 (20.9%)	77,830 (20.4%)	59,518 (19.9%)	39,604 (19.4%)	48,162 (18.0%)	< 0.001
Hypertension	222,974 (64.8%)	186,846 (61.0%)	225,132 (57.8%)	231,510 (55.6%)	205,451 (54.0%)	158,397 (53.2%)	107,310 (52.7%)	140,427 (52.7%)	< 0.001
SBP (mmHg)	128.9 ± 15.8	128.4 ± 15.6	128.5 ± 15.6	128.7 ± 15.7	129 ± 15.7	129.3 ± 15.8	129.6 ± 15.9	130.2 ± 16.3	< 0.001
DBP (mmHg)	78.7 ± 10.4	78.2 ± 10.1	78.5 ± 10.1	78.8 ± 10.2	79.1 ± 10.2	79.4 ± 10.2	79.7 ± 10.3	80.1 ± 10.4	< 0.001
Diabetes mellitus Duration, ≥ 5 years	132,768 (38.6%)	114,238 (37.3%)	133,056 (34.1%)	131,108 (31.5%)	110,505 (29.0%)	79,651 (26.7%)	49,963 (24.5%)	58,226 (21.8%)	< 0.001
Use of insulin	41,380 (12.0%)	33,397 (10.9%)	37,117 (9.5%)	35,125 (8.4%)	28,896 (7.6%)	20,834 (7.0%)	13,386 (6.5%)	17,806 (6.6%)	< 0.001
Use of OHA, ≥3	64,369 (18.7%)	52,905 (17.3%)	60,506 (15.5%)	59,298 (14.2%)	50,296 (13.2%)	36,981 (12.4%)	23,550 (11.5%)	29,402 (11.0%)	< 0.001
Fasting glucose (mg/dL)	142.4 ± 48.7	140.1 ± 45.3	141.8 ± 45.7	143.5 ± 45.5	145.1 ± 45.9	147.0 ± 46.5	148.9 ± 47.6	152.8 ± 51.2	< 0.001
Use of statin	177,777 (51.7%)	128,879 (42.1%)	119,496 (30.6%)	93,357 (22.4%)	72,261 (19.0%)	60,416 (20.2%)	50,766 (24.9%)	98,495 (37.0%)	< 0.001
Total cholesterol (mg/dL)	149.3 ± 35.5	162.3 ± 23.9	176.1 ± 22	190.3 ± 20.4	204.8 ± 19.5	219.9 ± 19.1	235.2 ± 18.8	268.3 ± 33.6	< 0.001
HDL-cholesterol (mg/dL)	51.6 ± 27.5	51.3 ± 21.6	51.5 ± 21.4	51.7 ± 21.6	51.7 ± 21.6	52.1 ± 23.0	52.6 ± 24.2	54.2 ± 30.2	< 0.001
LDL-cholesterol (mg/dL)	52.3 ± 15.6	77.4 ± 4.2	92.2 ± 4.3	106.9 ± 4.3	121.7 ± 4.3	136.6 ± 4.3	151.4 ± 4.2	185.2 ± 46.4	< 0.001
* Triglyceride (mg/dL)	174.5 (174.0-174.9)	142.4 (142.1-142.7)	140.1 (139.9-140.4)	139.9 (139.6-140.1)	140.5 (140.3-140.8)	142.5 (142.2-142.7)	144.4 (144.1-144.7)	151.4 (151.1-151.7)	< 0.001
* expressed as median (interquartile range)
BMI: body-mass-index; DBP: diastolic blood pressure; HDL: high-density lipoprotein; LDL: low-density lipoprotein; OHA: oral hypoglycemic agent; SBP: systolic blood pressure.

Impact Of LDL-cholesterol Level On SCA

Before adjustment of covariates, overall incidence rate (IR) of SCA categorized by LDL-cholesterol level was highest in group I (lowest LDL-cholesterol group; IR = 1.85; Table 2), followed by group II (IR = 1.58; Table 2), group III (IR = 1.51; Table 2), and group VIII (highest LDL-cholesterol group; IR = 1.41; Table 2). After adjustment of age and sex (Model 1), highest risk of SCA was observed in group I, followed by group VIII (Table 2 and Fig. 2). After adjustment of multiple covariates of sociodemographic and metabolic factors (Model 2), risk of SCA was highest in group VIII (adjusted hazard ratio [AHR] = 1.059, 95% CI = 1.007–1.113; p = 0.0203; Table 2 and Fig. 2), followed by group I (reference group; Table 2 and Fig. 2). Before consideration of covariates, overall IR reflected reverse J-shape relationship with LDL-cholesterol level, whereas adjustment of covariates showed U-shape association, with highest SCA risk in highest LDL-cholesterol group (group VIII). This finding was consistent across age subgroups (Supplementary Table S3).

Table 2

**Impact of LDL-cholesterol level on SCA and subgroup analysis**
LDL-cholesterol, mg/dL	N	Event	Duration	IR, per 1000	Adjusted HR
Whole cohort
< 70	343,781	4,313	2,335,003	1.847	1 (Ref.)
< 75	305,842	3,283	2,083,277	1.575	0.884 (0.844–0.925)
< 100	389,404	4,033	2,668,129	1.511	0.884 (0.846–0.923)
< 115	415,948	3,969	2,857,482	1.388	0.843 (0.806–0.881)
< 130	380,029	3,602	2,618,974	1.375	0.871 (0.832–0.912)
< 145	297,720	2,768	2,052,509	1.348	0.898 (0.855–0.943)
< 160	203,636	1,798	1,405,373	1.279	0.896 (0.847–0.947)
≥ 160	266,217	2,575	1,831,046	1.406	1.059 (1.007–1.113)
Male
< 70	235,091	3,420	1,587,304	2.154	1 (Ref.)
< 75	192,013	2,390	1,299,316	1.839	0.852 (0.809–0.898)
< 100	242,681	2,936	1,652,335	1.776	0.861 (0.819–0.905)
< 115	256,338	2,838	1,750,195	1.621	0.817 (0.777–0.860)
< 130	227,758	2,482	1,560,248	1.590	0.837 (0.794–0.883)
< 145	170,905	1,843	1,170,216	1.574	0.867 (0.819–0.919)
< 160	110,355	1,152	755,774	1.524	0.880 (0.822–0.941)
≥ 160	127,157	1,503	865,297	1.736	1.058 (0.995–1.125)
Female
< 70	108,690	893	747,698	1.194	1 (Ref.)
< 75	113,829	893	783,961	1.139	1.002 (0.913–1.100)
< 100	146,723	1,097	1,015,794	1.079	0.979 (0.895–1.070)
< 115	159,610	1,131	1,107,286	1.021	0.943 (0.862–1.032)
< 130	152,271	1,120	1,058,726	1.057	0.994 (0.908–1.088)
< 145	126,815	925	882,293	1.048	1.007 (0.916–1.107)
< 160	93,281	646	649,599	0.994	0.973 (0.877–1.079)
≥ 160	139,060	1,072	965,748	1.110	1.125 (1.028–1.232)
BMI < 25
< 70	179,832	2,746	1,198,668	2.290	1 (Ref.)
< 75	163,692	2,037	1,100,486	1.851	0.843 (0.796–0.893)
< 100	207,600	2,486	1,407,077	1.766	0.828 (0.784–0.875)
< 115	217,849	2,409	1,482,357	1.625	0.776 (0.734–0.821)
< 130	194,372	2,175	1,328,505	1.637	0.804 (0.759–0.852)
< 145	147,945	1,604	1,011,792	1.585	0.809 (0.759–0.861)
< 160	98,373	1,060	674,577	1.571	0.837 (0.779–0.899)
≥ 160	125,585	1,457	857,366	1.699	0.974 (0.913–1.039)
BMI ≥ 25
< 70	163,949	1,567	1,136,334	1.379	1 (Ref.)
< 75	142,150	1,246	982,791	1.267	0.952 (0.884–1.026)
< 100	181,804	1,547	1,261,051	1.226	0.978 (0.911–1.050)
< 115	198,099	1,560	1,375,125	1.134	0.954 (0.888–1.025)
< 130	185,657	1,427	1,290,468	1.105	0.979 (0.909–1.054)
< 145	149,775	1,164	1,040,716	1.118	1.043 (0.965–1.128)
< 160	105,263	738	730,795	1.009	0.987 (0.902–1.079)
≥ 160	140,632	1,118	973,679	1.148	1.194 (1.104–1.292)
Statin naïve
< 70	166,004	2,195	1,124,450	1.952	1 (Ref.)
< 75	176,963	1,906	1,200,642	1.587	0.810 (0.761–0.861)
< 100	269,908	2,744	1,845,959	1.486	0.781 (0.738–0.826)
< 115	322,591	2,959	2,213,922	1.336	0.729 (0.689–0.771)
< 130	307,768	2,816	2,117,839	1.329	0.758 (0.717–0.802)
< 145	237,304	2,127	1,633,582	1.302	0.784 (0.738–0.832)
< 160	152,870	1,316	1,051,772	1.251	0.794 (0.741–0.851)
≥ 160	167,722	1,617	1,148,656	1.407	0.963 (0.902–1.028)
Statin use
< 70	177,777	2,118	1,210,552	1.749	1 (Ref.)
< 75	128,879	1,377	882,635	1.560	0.954 (0.892–1.021)
< 100	119,496	1,289	822,169	1.567	1.020 (0.951–1.093)
< 115	93,357	1,010	643,559	1.569	1.062 (0.985–1.145)
< 130	72,261	786	501,135	1.568	1.109 (1.022–1.204)
< 145	60,416	641	418,926	1.530	1.127 (1.031–1.232)
< 160	50,766	482	353,600	1.363	1.060 (0.960–1.171)
≥ 160	98,495	958	682,389	1.403	1.198 (1.108–1.295)
Insulin naïve
< 70	302,401	3,272	2,071,675	1.579	1 (Ref.)
< 75	272,445	2,562	1,867,817	1.371	0.886 (0.842–0.934)
< 100	352,287	3,204	2,427,033	1.320	0.875 (0.833–0.920)
< 115	380,823	3,266	2,628,009	1.242	0.845 (0.804–0.888)
< 130	351,133	2,929	2,428,968	1.205	0.849 (0.806–0.893)
< 145	276,886	2,299	1,915,834	1.200	0.884 (0.837–0.934)
< 160	190,250	1,502	1,316,982	1.140	0.882 (0.829–0.939)
≥ 160	248,411	2,123	1,714,647	1.238	1.039 (0.982–1.098)
Insulin use
< 70	41,380	1,041	263,327	3.953	1 (Ref.)
< 75	33,397	721	215,460	3.346	0.874 (0.795–0.962)
< 100	37,117	829	241,096	3.438	0.918 (0.837–1.007)
< 115	35,125	703	229,473	3.063	0.826 (0.749–0.911)
< 130	28,896	673	190,006	3.541	0.978 (0.885–1.080)
< 145	20,834	469	136,675	3.431	0.963 (0.862–1.076)
< 160	13,386	296	88,390	3.348	0.971 (0.852–1.107)
≥ 160	17,806	452	1,16,398	3.883	1.169 (1.045–1.308)
LDL: low-density lipoprotein cholesterol; IR: incidence rate; HR, hazard ratio; BMI, body mass index.
Hazard ratio was adjusted with age, sex, income, body mass index, smoking status, alcohol consumption status, regular exercise, hypertension, fasting blood glucose, duration of diabetes mellitus, use of insulin, use of oral hypoglycemic agent, and use of statin.

Subgroup Analysis: Gender

The risk of SCA according to LDL-cholesterol level was evaluated in both gender (Table 2 and Fig. 3A). The overall incidence of SCA was significantly higher in male across all LDL-cholesterol subgroups. Both male and female showed similar patterns of SCA incidence, reflecting reverse J-shape curve. After adjustment of covariates, U-shape association was observed in male with highest and lowest LDL-cholesterol group having the higher risk of SCA (Table 2 and Fig. 3A). However, such U-shape association was significantly diluted in female (Table 2 and Fig. 3A).

Subgroup Analysis: Obesity

Comparison of obese group (BMI ≥ 25) and non-obese group (BMI < 25) showed markedly lower IR in obese group regardless of LDL-cholesterol (Table 2 and Fig. 3B). Multivariate model showed U-shape association between LDL-cholesterol level and SCA risk in non-obese group (Table 2 and Fig. 3B). However, such U-shape association was no longer present in the obese group.

Subgroup Analysis: Statin

Incidence of SCA regarding the use of statin did not differ significantly (Table 2 and Fig. 3C). In statin-naïve group, adjustment of covariates resulted in prominent U-shape curve, with highest risk of SCA in lowest LDL-cholesterol group (group I; reference group), followed by highest LDL-cholesterol group (group VIII; AHR = 0.963; 95% CI = 0.902–1.028; Table 2 and Fig. 3C). The risk of SCA in LDL-cholesterol group II to VII was significantly lower than group I. In contrast, the association between LDL-cholesterol and SCA risk was rather linear association in the statin-use group with the highest SCA risk observed in the highest LDL-cholesterol group (AHR = 1.198; 95% CI = 1.108–1.295; Table 2 and Fig. 3C).

Subgroup Analysis: Insulin

Incidence of SCA was significantly higher in diabetes people using insulin compared to insulin-naïve diabetes people, which was consistent across all LDL-cholesterol level (Table 2 and Fig. 3D). In the insulin-naïve group, adjusted-HR was significantly lower in LDL-cholesterol group II to VII and numerically higher in group VIII as compared with the reference group (group I). In the insulin-use group, significantly higher risk of SCA was observed in the highest LDL-cholesterol group (AHR = 1.169; 95% CI = 1.045–1.308; Table 2 and Fig. 3D).

This study investigated the association of dyslipidemia, in which its severity categorized by LDL-cholesterol level, with the risk of SCA in diabetes patients. Incidence of SCA reflected reverse J-shape association with LDL-cholesterol level: SCA was most prevalent in the lowest LDL-cholesterol group (< 70 mg/dL), and decreased as LDL-cholesterol level rises, until LDL-cholesterol level reached 160 mg/dL. After adjustment of various confounding factors, lowest LDL-cholesterol group (< 70 mg/dL) resulted second highest risk of SCA, following the highest LDL-cholesterol group (≥ 160 mg/dL) resulting in a U-shape association. LDL-cholesterol level resulted reverse relationship with adjusted risk of SCA in lower LDL-cholesterol population, till it reached 115mg/dL, which was contrary finding from traditional association of LDL-cholesterol with cardiovascular disease.

This study shows strength in investigating yet undiscovered association of SCA and LDL-cholesterol level in diabetes population. Analysis was based on nationwide health insurance database that include large volume of diabetes population, and various patient-related factors were adjusted from vast store of database on sociodemographic information as well as laboratory markers and medication use.

Sudden Cardiac Arrest, Cardiovascular Disease, And Dyslipidemia

SCA features an abrupt, unexpected onset of cardiac arrest, which may lead to irreversible sequalae even after prompt, successful resuscitation. Primary prevention of SCA with identification and stratification of risk factors have been of concern for decades, but is an area of challenge, due to dynamic course of SCA entangled with various risk factors that influence each other. Dyslipidemia measured with serum LDL-cholesterol level has been identified as a major risk factor for cardiovascular events. Traditionally, LDL-cholesterol level has reflected a positive linear correlation with cardiovascular events.[20, 21] In recent study, lowering of LDL-cholesterol level as low as 40mg/dL has resulted additional reduction of major cardiovascular events emphasizing reduction of LDL-cholesterol level as low as possible.[9, 21] Moreover, prolonged lowering of LDL-cholesterol is associated with lower risk of atherosclerotic cardiovascular disease.[22] Accordingly, “the lower, the better” paradigm has introduced potent lipid lowering strategies in clinical field, including ezetimibe and proprotein convertase subtilisin/Kexin 9 inhibitors in addition to statin therapy.

Previous Studies On Association Of Ldl-cholesterol And Sudden Cardiac Arrest

In this study, the traditional effect of LDL-cholesterol on cardiovascular mortality was reversed in terms of SCA risk: lowest LDL-cholesterol group showed significantly increased risk of SCA. This reverse association of low LDL-cholesterol level with SCA was even more emphasized in statin-naïve subgroup and non-obese subgroup. Several studies have investigated associations of LDL-cholesterol with the risk of SCA, in which most of them did not find any significant association between LDL-cholesterol and SCA.[11, 14] Hosadurg et al. have reported similar findings with our study: compared with control cohort, out-of-hospital sudden unexpected death cases in North Carolina had significantly lower level of mean total cholesterol, non-high-density lipoprotein cholesterol, and notably, LDL-cholesterol.[12] It had introduced a novel finding of reverse association of low LDL-cholesterol with risk of SCA, but had limitation of small sized samples (n = 399) from geographically limited area, with relatively high proportion of missing values (more than 30%). Our study has further focused on diabetes population, that are more susceptible to dyslipidemia and cardiovascular disease, and confirmed the concept based on large cohort. Prospective nature of the cohort with demonstration of chronological association between LDL-cholesterol level and SCA risk is another strong point of this study.

Possible Mechanisms For Reverse Association

Although diabetes mellitus and dyslipidemia are known to be independent risk factors for cardiovascular event, it is a cluster of plasma lipid and lipoprotein abnormalities that are metabolically interrelated, known as diabetic dyslipidemia.[8] Therefore, more strict control of lipid level is recommended for prevention of further cardiovascular event in diabetes population. Nevertheless, our study has suggested that low LDL-cholesterol can be associated with paradoxically increased risk of SCA.

The exact mechanisms of the reversal of SCA risk in low LDL-cholesterol group are not established, but it might be supported by several explanations. First, subpopulation of low LDL-cholesterol level (< 70 mg/dL) might represent high risk group for SCA. Patients with pre-existing severe systemic condition (i.e, malnutrition, respiratory disease, inflammatory disease, or malignancy) that is more susceptible to SCA may exhibit low LDL-cholesterol level as a secondary consequence.[23] Loss of association between low LDL-cholesterol level and increased risk of SCA in people taking statins also support the hypothesis that low LDL-cholesterol is a surrogate marker for SCA and not a direct determinant. Second possible mechanism is the protective effect of high serum cholesterol on immune system. Several studies have suggested that serum cholesterol plays protective role on bacterial and viral infection by various mechanisms, such as binding to endotoxin, and increase of lymphocytes.[24–28] This protective effect of cholesterol may be more pronounced on diabetes population, since they are more vulnerable to systemic infection that might cause major organ dysfunction as well as death. The immunomodulatory function of cholesterol also affect the development of virus-related cancer, which may also be related with death.[29] However, this cannot fully explain the phenomenon, since SCA event was confined to out-of-hospital-cardiac-arrest claimed at emergency room, which excludes majority of cancer related death and pre-existing infection. Lastly, undiscovered genetic susceptibility that causes both low LDL-cholesterol and SCA might lie in diabetes patients which needs to be further investigated.

Limitations

There are several limitations in this study. First, although SCA risk was adjusted with multiple covariates in our multivariate model which was validated in our previous studies, there can be residual confounders. Second, although the severity of diabetes mellitus was adjusted through prescription of oral hypoglycemic agents or insulin and fasting blood glucose, HbA1c level was not applied for multivariate adjustment. Further encompassment of multifactorial conditions related to diabetes mellitus and dyslipidemia might provide more comprehensive understanding of this reversal of SCA risk in low LDL-cholesterol group. Third, participants included in our study might not represent generalized diabetes population. This study is limited to East Asian population, exclusively confined to South Korean citizen. Since LDL-cholesterol level vary significantly depending on ethnic group, the association between LDL-cholesterol and SCA risk might differ in other populations. Fourth, temporal change in LDL-cholesterol level was not evaluated in this study. Risk of cardiovascular event is influenced not only by the level of LDL-cholesterol but also temporal exposure to dyslipidemia.[9, 30] Association between temporal change in LCL-C and risk of SCA will be an area of future research.

In summary, in diabetes patients, not only high LDL-cholesterol but also low LDL-cholesterol was associated with increased risk of SCA. Although cause and effect relationship cannot be established based on this study, low LDL-cholesterol can be a surrogate marker for high risk group for SCA among diabetes people. In order to apply this finding to clinical field and preventive measures for SCA, mechanisms for this reversed relationship of LDL-cholesterol and SCA needs to be further investigated.

AHR

adjusted hazard ratio

confidence interval

ICD-10

International Classification of Disease, 10th revision

incidence rate

K-NHIS

Korean National Health Insurance Service

SCA

sudden cardiac arrest

Ethics approval and consent to participate

The Institutional Review Board of Korea University Medicine Anam Hospital and official review committee of the K-NHIS approved this specific study (IRB No.: 2021AN0185). The requirement for written informed consent was waived by the Institutional Review Board of Korea University Medicine Anam Hospital, due to the retrospective nature of the study.

Consent for publication

Not applicable

Availability of data and materials

The data underlying this article are available in the article. The raw data underlying this article cannot be shared publicly due to privacy reasons and legal regulations of Republic of Korea. The raw data is stored and analyzed only in the designated server managed by the K-NHIS.

Competing interests

Jong-Il Choi reports honoraria from Daiichi Sankyo, Inc.; Boehringer-Ingelheim; Abbott; Sanofi Genzyme; Samjin pharma; Yuhan; Chong Keun Dang, Hanmi Pharmaceutical and research grants from Sanofi Genzyme; Medtronic; Chong Keun Dang.

Funding

This work was supported by a Korea University grant (J-I.C), a grant from Korea University Anam Hospital, Seoul, Republic of Korea (J-I.C), and in part by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT, Ministry of Science and ICT) (No. 2021R1A2C2011325 to J-I.C). The funders had no role in data collection, analysis, or interpretation; trial design; patient recruitment; or any other aspect pertinent to the study.

Authors’ contributions

J-I Choi had full access to all data in this study and takes responsibility for its integrity and analytical accuracy. The study concept and design were developed by YG Kim, JH Jeong, YY Choi, HS Lee, K-D Han, J-I Choi, and YH Kim. The data analysis and interpretation were performed by YG Kim, JH Jeong, YY Choi, HS Lee, K-D Han, K Min, and J-I Choi. The manuscript was drafted by YG Kim, JH Jeong, K-D Han, and J-I Choi. The statistical analysis was performed by YG Kim, K-D Han, and J-I Choi. The data collection was performed by YG Kim, S-Y Roh, KD Han, JH Jeong, J Shim, and J-I Choi.

Acknowledgements

None

Al-Khatib SM, Stevenson WG, Ackerman MJ, Bryant WJ, Callans DJ, Curtis AB, et al. 2017 AHA/ACC/HRS Guideline for Management of Patients With Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2018;72:1677-749.
Priori SG, Blomstrom-Lundqvist C, Mazzanti A, Blom N, Borggrefe M, Camm J, et al. 2015 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death: The Task Force for the Management of Patients with Ventricular Arrhythmias and the Prevention of Sudden Cardiac Death of the European Society of Cardiology (ESC). Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC). Eur Heart J. 2015;36:2793-867.
Myerburg RJ, Kessler KM, Castellanos A. Sudden cardiac death. Structure, function, and time-dependence of risk. Circulation. 1992;85:I2-10.
Organization WH, Norrving B, Puska P, Mendis S, World Health O, World Stroke O, et al. Global Atlas on Cardiovascular Disease Prevention and Control. Geneva: World Health Organization; 2011.
Chugh SS. Early identification of risk factors for sudden cardiac death. Nat Rev Cardiol. 2010;7:318-26.
Kim YG, Roh SY, Han K-D, Jeong JH, Choi YY, Min K, et al. Hypertension and diabetes including their earlier stage are associated with increased risk of sudden cardiac arrest. Scientific Reports. 2022;12.
Kim YG, Han K, Jeong JH, Roh S-Y, Choi YY, Min K, et al. Metabolic Syndrome, Gamma-Glutamyl Transferase, and Risk of Sudden Cardiac Death. Journal of Clinical Medicine. 2022;11:1781.
Visseren FLJ, Mach F, Smulders YM, Carballo D, Koskinas KC, Back M, et al. 2021 ESC Guidelines on cardiovascular disease prevention in clinical practice. Eur Heart J. 2021;42:3227-337.
Mach F, Baigent C, Catapano AL, Koskinas KC, Casula M, Badimon L, et al. 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J. 2020;41:111-88.
Jouven X, Desnos M, Guerot C, Ducimetiere P. Predicting sudden death in the population: the Paris Prospective Study I. Circulation. 1999;99:1978-83.
Albert CM, Ma J, Rifai N, Stampfer MJ, Ridker PM. Prospective study of C-reactive protein, homocysteine, and plasma lipid levels as predictors of sudden cardiac death. Circulation. 2002;105:2595-9.
Hosadurg N, Bogle BM, Joodi G, Sadaf MI, Pursell I, Mendys PM, et al. Lipid Profiles in Out-of-Hospital Sudden Unexpected Death. Mayo Clin Proc Innov Qual Outcomes. 2018;2:257-66.
Jouven X, Lemaitre RN, Rea TD, Sotoodehnia N, Empana JP, Siscovick DS. Diabetes, glucose level, and risk of sudden cardiac death. Eur Heart J. 2005;26:2142-7.
Kurl S, Laaksonen DE, Jae SY, Makikallio TH, Zaccardi F, Kauhanen J, et al. Metabolic syndrome and the risk of sudden cardiac death in middle-aged men. Int J Cardiol. 2016;203:792-7.
Kim YG, Han KD, Choi JI, Choi YY, Choi HY, Boo KY, et al. Non-genetic risk factors for atrial fibrillation are equally important in both young and old age: A nationwide population-based study. Eur J Prev Cardiol. 2021;28:666-76.
Kim YG, Han KD, Kim DY, Choi YY, Choi HY, Roh SY, et al. Different Influence of Blood Pressure on New-Onset Atrial Fibrillation in Pre- and Postmenopausal Women: A Nationwide Population-Based Study. Hypertension. 2021;77:1500-9.
Kim YG, Han KD, Choi JI, Choi YY, Choi HY, Shim J, et al. Premature ventricular contraction is associated with increased risk of atrial fibrillation: a nationwide population-based study. Sci Rep. 2021;11:1601.
Roh SY, Choi JI, Kim MS, Cho EY, Kim YG, Lee KN, et al. Incidence and etiology of sudden cardiac arrest in Koreans: A cohort from the national health insurance service database. PLoS One. 2020;15:e0242799.
Kim YG, Oh SK, Choi HY, Choi JI. Inherited arrhythmia syndrome predisposing to sudden cardiac death. Korean J Intern Med. 2021;36:527-38.
Silverman MG, Ference BA, Im K, Wiviott SD, Giugliano RP, Grundy SM, et al. Association Between Lowering LDL-C and Cardiovascular Risk Reduction Among Different Therapeutic Interventions A Systematic Review and Meta-analysis. Jama-Journal of the American Medical Association. 2016;316:1289-97.
Sabatine MS, Giugliano RP, Keech AC, Honarpour N, Wiviott SD, Murphy SA, et al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease. N Engl J Med. 2017;376:1713-22.
Ference BA, Ginsberg HN, Graham I, Ray KK, Packard CJ, Bruckert E, et al. Low-density lipoproteins cause atherosclerotic cardiovascular disease. 1. Evidence from genetic, epidemiologic, and clinical studies. A consensus statement from the European Atherosclerosis Society Consensus Panel. European Heart Journal. 2017;38:2459-72.
Ravnskov U, Diamond DM, Hama R, Hamazaki T, Hammarskjold B, Hynes N, et al. Lack of an association or an inverse association between low-density-lipoprotein cholesterol and mortality in the elderly: a systematic review. Bmj Open. 2016;6.
Ravnskov U. High cholesterol may protect against infections and atherosclerosis. Qjm-an International Journal of Medicine. 2003;96:927-34.
Cavaillon JM, Fitting C, Haeffnercavaillon N, Kirsch SJ, Warren HS. Cytokine Response by Monocytes and Macrophages to Free and Lipoprotein-Bound Lipopolysaccharide. Infection and Immunity. 1990;58:2375-82.
Weinstock C, Ullrich H, Hohe R, Berg A, Baumstark MW, Frey I, et al. Low-Density Lipoproteins Inhibit Endotoxin Activation of Monocytes. Arteriosclerosis and Thrombosis. 1992;12:341-7.
Losche W, Krause S, Pohl A, Pohl C, Liebrenz A, Schauer I, et al. Functional-Behavior of Mononuclear Blood-Cells from Patients with Hypercholesterolemia. Thrombosis Research. 1992;65:337-42.
Muldoon MF, Marsland A, Flory JD, Rabin BS, Whiteside TL, Manuck SB. Immune system differences in men with hypo- or hypercholesterolemia. Clinical Immunology and Immunopathology. 1997;84:145-9.
Read SA, Douglas MW. Virus induced inflammation and cancer development. Cancer Letters. 2014;345:174-81.
Ference BA, Yoo W, Alesh I, Mahajan N, Mirowska KK, Mewada A, et al. Effect of Long-Term Exposure to Lower Low-Density Lipoprotein Cholesterol Beginning Early in Life on the Risk of Coronary Heart Disease A Mendelian Randomization Analysis. Journal of the American College of Cardiology. 2012;60:2631-9.

No competing interests reported.

Supplementarymaterial20230110.docx

Download PDF

Journal Publication

published 20 Feb, 2023

Read the published version in Cardiovascular Diabetology →

Editorial decision: Major revision
30 Jan, 2023
Reviews received at journal
21 Jan, 2023
Reviewers agreed at journal
20 Jan, 2023
Reviewers invited by journal
20 Jan, 2023
Editor assigned by journal
20 Jan, 2023
Submission checks completed at journal
20 Jan, 2023
First submitted to journal
16 Jan, 2023

You are reading this latest preprint version

Association between Low-density Lipoprotein Cholesterol and Sudden Cardiac Arrest in People with Diabetes Mellitus

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Research Design And Methods

Database

Study Population

Definitions Of Variables

Statistical analysis

Results

Study population

Impact Of LDL-cholesterol Level On SCA

Subgroup Analysis: Gender

Subgroup Analysis: Obesity

Subgroup Analysis: Statin

Subgroup Analysis: Insulin

Discussion

Sudden Cardiac Arrest, Cardiovascular Disease, And Dyslipidemia

Previous Studies On Association Of Ldl-cholesterol And Sudden Cardiac Arrest

Possible Mechanisms For Reverse Association

Limitations

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1