Sage Journals: Discover world-class research

Abstract

Introduction

Low-density lipoprotein cholesterol (LDL) is a known contributing factor to atherosclerotic cardiovascular disease (ASCVD) and a primary therapeutic target for medical management of ASCVD. Non-high-density lipoprotein cholesterol (non-HDL) has recently been identified as a secondary therapeutic target but is not yet widely used in vascular surgery patients. We sought to assess if vascular surgery patients were undertreated per non-HDL therapeutic guidelines.

Methods

This was an observational study that used a single-center database to identify a cohort of adult patients who received care from a vascular surgery provider from 01/2001 to 07/2021. ICD-9/10-CM codes were used to identify patients with a medical history of hyperlipidemia (HLD), coronary artery disease (CAD), cerebrovascular occlusive disease (CVOD), peripheral artery disease (PAD), hypertension (HTN), or diabetes mellitus (DM). Patient smoking status and medications were also identified. Lab values were obtained from the first and last patient encounter within our system. Primary outcomes were serum concentrations of LDL and non-HDL, with therapeutic thresholds defined as 70 mg/dL and 100 mg/dL, respectively.

Results

The cohort included 2465 patients. At first encounter, average age was 59.3 years old, 21.4% were on statins, 8.4% were on a high-intensity statin, 25.7% were diagnosed with HLD, 5.2% with CAD, 15.3% with PAD, 26.3% with DM, 18.6% with HTN, and 2.1% with CVOD. At final encounter, mean age was 64.8 years, 23.5% were on statins with 10.1% on high-intensity statin. Diagnoses frequency did not change at final encounter. At first encounter, nearly two-thirds of patients were not at an LDL <70 mg/dL (62.3%) or non-HDL <100 mg/dL (66.0%) with improvement at final encounter to 45.2 and 40.5% of patients not at these LDL or non-HDL treatment thresholds, respectively. Patients on statins exhibited similar trends with 51.1 and 50.1% of patients not at LDL or non-HDL treatment thresholds at first encounter and 39.9 and 35.4% not at LDL or non-HDL treatment thresholds at last encounter. Importantly, 6.9% of patients were at LDL but not non-HDL treatment thresholds.

Discussion

Among vascular surgery patients, over half did not meet non-HDL targets. These results suggest that we may be vastly under-performing adequate medical optimization with only about one-fourth of patients on a statin at their final encounter and approximately one-tenth of patients being treated with a high-intensity statin. With recent evidence supporting non-HDL as a valuable measurement for atherosclerotic risk, there is potential to optimize medical management beyond current high-intensity statin therapy. Further investigation is needed regarding the risk of adverse events between patients treated with these varied therapeutic targets.

Keywords

Optimal medical therapy vascular medicine LDL non-HDL

Introduction

Atherosclerotic cardiovascular disease (ASCVD) is defined as acute coronary syndrome, myocardial infarction, angina, stroke or transient ischemic attack (TIA) due to cerebrovascular occlusive disease (CVOD), and/or peripheral artery disease (PAD) and is the leading cause of death in the United States.^1–3 Atherogenesis is driven by several factors, especially by atherogenic lipoproteins which play fundamental roles in the initiation and evolution of atherosclerosis and ASCVD.² Preventive measures via lifestyle modifications and risk factor modification with antiplatelet agents and statin therapy are critical for primary and secondary prevention of ASCVD.⁴ Low-density lipoprotein (LDL) cholesterol has historically been the primary medical target for primary and secondary prevention of ASCVD with a target value of less than 100 mg/dL.^4–8 However, current guidelines recommend considering ASCVD risk enhancers that include lipid and biomarkers such as elevated triglycerides rather than targeting specific LDL values alone.^5,9,10 Moreover, a growing body of literature has highlighted the importance of measuring and modifying other components of the lipid profile that may provide more accurate risk assessment, chiefly non-high-density lipoprotein (non-HDL) cholesterol and apolipoprotein B (Apo B).^11–14

Non-HDL defines the cholesterol content of all atherogenic lipoproteins and can be measured by subtracting the HDL from the total cholesterol (TC).¹⁵ Non-HDL is a better marker for risk of coronary artery disease and is more closely associated with coronary atheroma progression than LDL.^12,15,16 Furthermore, non-HDL has been shown to be a reliable predictor of adverse cardiovascular outcomes amongst patients on lipid-lowering therapy and to be as good or better than LDL in predicting ASCVD risk.^17,18 Non-HDL has been shown to be independently associated with both CVOD and PAD.^19–22 In fact, non-HDL has been added as secondary treatment target in patients with elevated triglycerides with a treatment threshold value of non-HDL 30 mg/dL above the LDL treatment threshold (i.e., non-HDL threshold of 100 mg/dL when the guideline-defined LDL threshold is 70 mg/dL).^4,22–25

Despite its association with PAD and CVOD, non-HDL is still not routinely used as a treatment target nor risk predictor in vascular surgery patients, who are in general high risk for ASCVD. High-intensity statins remain the mainstay of treatment for vascular surgery patients with PAD and/or CVOD with current treatment thresholds remaining unclear.^4,24–26 It is unclear how many vascular surgery patients have elevated non-HDL and thus may not be medically optimized and/or undertreated with statin therapy. Thus, we studied this high-risk population of vascular surgery patients to examine frequency of elevated non-HDL levels in addition to those who are not at proposed LDL and/or non-HDL treatment thresholds. In addition, we aimed to investigate whether patients with elevated non-HDL levels had higher proportions of PAD and/or CVOD.

Methods

This study was an observational cohort study of adults treated by a vascular surgery provider at the University of Michigan (U-M) from January 2001 to July 2021. Patients were selected from DataDirect, a single-center dataset using diagnosis codes from the International Classification of Disease 9^th and 10^th edition (ICD9/ICD10). DataDirect is a self-serve tool that enables access to clinical data such as diagnoses, encounters, procedures, medications, and labs from more than four million surgical patients from across the U-M Health System. The Institutional Review Board at University of Michigan approved this study and determined that informed consent was waived since it was a secondary analysis of a de-identified database.

Study population

The study population consisted of adult surgery patients and/or recipients of anesthesia who received care at the U-M in an inpatient and/or outpatient setting from a vascular surgery provider from 01/2001 to 07/2021. DataDirect catalogs information from all surgery patients and/or recipients of anesthesia. Patients were not included or excluded based on surgical procedures or providers but rather included on inpatient and/or outpatient encounters with vascular surgery providers. Patients were included if they had laboratory values and/or medication information in the database. Patient encounters were included if they were associated with lipid focused lab values: Triglycerides (TG), HDL, LDL, non-HDL, or TC. First encounter was defined as the first inpatient or outpatient encounter with a vascular surgery provider that was associated with lipid focused labs, and last encounter was defined as the last inpatient or outpatient encounter by a vascular surgery provider associated with lipid labs. Patients without more than one encounter associated with lipid-focused labs were excluded (Figure 1). ICD9/ICD10 codes were used to identify diagnosis of hyperlipidemia (HLD), CAD, CVOD, PAD, hypertension (HTN), and/or diabetes mellitus (DM) (Supplemental Table 1). General demographic information, statin usage, and lipid-focused lab values were collected. LDL is calculated at the U-M using the Friedewald equation.²⁷ We also calculated LDL using the Sampson²⁸ and the Martin-Hopkins equations.²⁹ Non-HDL was calculated using total cholesterol minus HDL cholesterol.

Figure 1.

Adult surgical patients seen by a vascular surgery provider at the University of Michigan between 2001 and 2021 were identified. Patients were included if they had patient encounters that were associated with medications, diagnoses, and lipid-focused laboratory values (TG, HDL, LDL, non-HDL, and/or TC). Patients were included if they have lipid-focused lab values at more than one patient encounter.

Data analysis

Diagnoses and lab values were assessed at the patients’ first encounter and last encounter and at any encounter within the database. The first encounter was defined as the earliest patient encounter that was associated with laboratory measurements of triglycerides, HDL, LDL, non-HDL, and/or TC while the last encounter was defined as the latest patient encounter associated with the above laboratory measurements. We first assessed average age, statin use, and encounter diagnoses between first and last patient encounters. Statin use was defined as documentation of any medication within the statin class during an encounter. Encounter diagnoses were defined using the aforementioned IDC9/ICD10 codes. We then compared the aforementioned lipid-focused lab values and diagnoses between first and last encounter. Primary outcomes were serum concentrations of LDL (calculated via the Friedewald equation) and non-HDL, with therapeutic thresholds of 70 mg/dL and 100 mg/dL, respectively. Additionally, patients were divided into cohorts based on their LDL (calculated via the Friedewald, Sampson, and Martin-Hopkins equations) and non-HDL lab values and therapeutic thresholds (LDL < or ≥70 mg/dL, non-HDL < or ≥100 mg/dL, LDL <70 mg/dL and non-HDL <100 mg/dL, and LDL <70 mg/dL but non-HDL >100 mg/dL). We then compared encounter diagnoses at last patient encounter for patients who were within these therapeutic thresholds.

Statistical analysis was done with Stata (StataCorp, College Station, TX). Paired student’s T test was used to compare average lab values and numerical data between groups. Chi squared analysis was used to compare average diagnoses and categorical data between patient cohorts.

Results

In total, 2465 patients were included (Figure 1). Approximately 50.3% (1241) of patients were male, while 25.6% (630) were female and 24.1% (594) had an unknown and/or unreported gender. The majority of patients (2014, 81.7%) were Caucasian, while 13.9% (343) were African American, 1.9% (48) reported a race of other, and 2.4% (60) had an unknown race. Approximately 93.3% (2300) of patients were non-Hispanic or Latino, 2.2% (55) were Hispanic or Latino, and 4.5% (110) had an unknown ethnicity. Regarding smoking status, 55.2% (1360) reported former or current tobacco use, 36.6% (902) reported no previous use of tobacco, and 8.2% (203) had unknown tobacco use history (Table 1).

Table 1.

Patient demographics, statin use, and diagnoses at first and final encounters.

	First encounter	Final encounter
Total patients	2465	—
Race
Caucasian	2014 (81.7%)	—
African American	343 (13.9%)	—
Other	48 (1.9%)	—
Unknown	60 (2.4%)	—
Ethnicity
Hispanic or Latino	55 (2.2%)	—
Non-Hispanic or Latino	2300 (93.3%)	—
Unknown	110 (4.5%)	—
Gender
Male	1241 (50.3%)	—
Female	630 (25.6%)	—
Unknown	594 (24.1%)	—
Smoking status
Former/current	1360 (55.2%)	—
Never	902 (36.6%)	—
Unknown	203 (8.2%)	—
Average age	59.3	64.8
Statin	528 (21.4%)	578 (23.5%)
High-intensity statin	206 (8.4%)	248 (10.1%)
Atorvastatin 40 mg	102 (19.0%)	119 (20.3%)
Atorvastatin 80 mg	93 (17.3%)	118 (20.1%)
Rosuvastatin 20 mg	7 (1.3%)	6 (1.2%)
Rosuvastatin 40 mg	4 (0.7%)	5 (0.9%)
Diagnoses
Missing diagnoses	633 (25.7%)	369 (15.0%)
Hyperlipidemia	579 (31.6%)	704 (33.6%)
Coronary artery disease	95 (5.2%)	133 (6.3%)
Peripheral artery disease	280 (15.3%)	413 (19.7%)
Diabetes	481 (26.3%)	305 (14.6%)
Hypertension	340 (18.6%)	487 (23.2%)
Cerebrovascular occlusive disease	38 (2.1%)	74 (3.5%)

At first encounter, patients had an average age of 59.3 years old. At their last encounter, the average patient age was 64.8 years old. Only 21.4% (528) of patients were on a statin medication at first encounter with an increase to 23.5% (578) of patients on statins at their last encounter. Only 8.4% (206) of patients were on a high-intensity statin at their first encounter with an increase to 10.1% (248) of patients on a high-intensity statin at their last encounter. Of patients on statins at their first encounter, 19.0% (102) were on atorvastatin 40 mg, 17.3% (93) were on atorvastatin 80 mg, 1.3% (7) were on rosuvastatin 20 mg, and 0.7% (4) were on rosuvastatin 40 mg. Of patients on statins at their last encounter, 20.3% (119) were on atorvastatin 40 mg, 20.1% (118) were on atorvastatin 80 mg, 1.2% (6) were on rosuvastatin 20 mg, and 0.9% (5) were on rosuvastatin 40 mg. Approximately 25.7% of patients did not have diagnosis information at their first encounter and 15.0% of patients did not have diagnosis information at their last encounter. At first encounter, 31.6% (579) patients carried a diagnosis of HLD, 5.2% (95) had CAD, 15.3% (280) had PAD, 26.3% (481) had DM, 18.6% (340) had HTN, and 2.1% (38) had CVOD. At last encounter, 33.6% (704) carried a diagnosis of HLD, 6.3% (133) had CAD, 19.7% (413) had PAD, 14.6% (305) had DM, 23.2% (487) had HTN, and 3.5% (74) had CVOD (Table 1). Male patients at initial (67.6 vs 45.9%, p < 0.001) and final (63.9 vs 46.3%, p < 0.001) encounters as well as older patients at initial (62.4 vs 58.5, p < 0.001) and final (66.7 vs 64.2, p < 0.001) encounters had higher percentages of statin use (Table 2). Patients with a diagnosis of CAD had the highest percentage of statin use at first encounter (51.3%) while those with HLD had the lowest (29.0%). Patients who were not on statins had higher percentages of HLD (71.0% vs 29.0%), PAD (64.2% vs 36.8%), DM (70.0% vs 39.3%), HTN (64.9% vs 35.1%), and CVOD (58.2% vs 41.8%) at first encounter (p < 0.001). Patients who were not on statins had higher percentages of PAD (57.9% vs 42.1%), DM (60.0% vs 40.0%), and HTN (52.8% vs 47.2%) at their final encounter (p < 0.0 01). Interestingly, patients not at statins had lower percentages of CAD at first encounter (48.7% vs 51.3%, p < 0.001) and last encounter (45.5% vs 54.5%, p < 0.001), lower percentages of CVOD (47.0% vs 53.0%, p < 0.001) at final encounter, and no difference in diagnosis of HLD at final encounter (Table 2). Statin usage appeared to increase when comparing first to final encounter across all diagnoses (Table 2).

Table 2.

Patients treated with and without statins.

	Initial encounter			Final encounter
Number	Statin 528	No statin 1925	p-value	Statin 579	No statin 1885	p-value
Age	62.4	58.5	<0.001	66.7	64.2	0.007
Race
Caucasian	432 (81.8%)	1582 (81.2%)	0.972	472 (81.5%)	1543 (81.9%)	0.765
African American	74 (14.0%)	269 (14.0%)	—	82 (14.2%)	261 (13.8%)	—
Asian	7 (1.3%)	31 (1.6%)	—	6 (1.0%)	32 (1.7%)	—
American Indian/Alaskan Native	4 (0.8%)	11 (0.6%)	—	4 (0.7%)	0 (0.0%)	—
Ethnicity
Non-Hispanic or Latino	511 (96.8%)	1789 (92.9%)	<0.001	544 (94.0%)	1757 (93.2%)	0.756
Hispanic or Latino	8 (1.5%)	47 (2.4%)	—	11 (1.9%)	44 (2.3%)	—
Gender
Male	357 (67.6%)	884 (45.9%)	<0.001	370 (63.9%)	872 (46.3%)	<0.001
Female	169 (32.0%)	461 (23.9%)	—	205 (35.4%)	425 (22.5%)	—
Diagnoses
Hyperlipidemia	158 (29.0%)	388 (71.0%)	<0.001	12 (31.6%)	26 (68.4%)	0.233
Coronary artery disease	297 (51.3%)	282 (48.7%)	<0.001	384 (54.5%)	320 (45.5%)	<0.001
Peripheral artery disease	34 (36.8%)	61 (64.2%)	<0.001	56 (42.1%)	77 (57.9%)	<0.001
Diabetes	82 (29.3%)	198 (70.7%)	<0.001	165 (40.0%)	248 (60.0%)	<0.001
Hypertension	169 (35.1%)	312 (64.9%)	<0.001	144 (47.2%)	161 (52.8%)	<0.001
Cerebrovascular occlusive disease	142 (41.8%)	198 (58.2%)	<0.001	258 (53.0%)	229 (47.0%)	<0.001

Paired T test used to compare average age between patients on statins versus those without statins at initial and final encounters. Chi squared analysis was used to compare race, ethnicity, and gender between patients on statins versus those not on statins at initial and final encounters.

When comparing initial and final encounters for all patients, there was a decrease in average TG (171.5 vs 159.2, p < 0.001), TC (168.0 vs 149.6, p < 0.001), HDL (35.0 vs 33.0, p < 0.001), LDL (69.9 vs 54.0, p < 0.001), and non-HDL (113.4 vs 78.6, p < 0.001). At initial encounter, 62.3% of patients were not at LDL of <70 mg/dL and 66.0% of patients were not at a non-HDL of <100 mg/dL. This improved with 45.2% of patients not at a LDL of <70 mg/dL and 40.5% of patients not at a non-HDL of <100 mg/dL. Patients who were on statins demonstrated a significant decrease in average TC (123.4 vs 100.1, p < 0.001), HDL (33.0 vs 29.0, p < 0.001), LDL (66.5 vs 49.7, p < 0.001), and non-HDL (90.7 vs 71.6, p < 0.001) when comparing initial and final encounters. At initial encounter, 51.1% of patients on statins were not at LDL of <70 mg/dL and 50.1% of patients on statins were not at a non-HDL of <100 mg/dL. This improved with 39.9% of patients not at LDL of <70 mg/dL and 35.4% of patients were not at non-HDL of <100 mg/dL. At final encounter, 36.3% were at LDL of <70 mg/dL; however, 14.1% of those patients remained above a non-HDL of >100 mg/dL. Patients on statins exhibited an overall trend of lower triglycerides, cholesterol, HDL, LDL, and non-HDL. Additionally, more patients taking a statin were at a non-HDL of <100 mg/dL and LDL of <70 mg/dL (Table 3). There were no significant differences between average LDL or non-HDL values between races at initial or final encounter (Supplemental Table 2). Additionally, there were no significant differences between statin usage, average TC, triglycerides, HDL, LDL, or non-HDL between patients who were followed for more than 3 years compared to those followed less than 3 years (Supplemental Table 3).

Table 3.

Patient lab values at initial and final encounters.

	All patient initial encounter			All patient final encounter			p-value*	On statin initial encounter			On statin final encounter			p-value**
	N (%)	Average (SD)	N > goal	N (%)	Average (SD)	N > goal	p-value*	N	Average (SD)	N > goal	N	Average (SD)	N > goal	p-value**
Triglycerides	2193 (90.0)	171.5 (136.6)	—	2072 (84.1)	159.2 (129.8)	—	<0.001	521 (98.7)	142.3 (106.1)	—	499 (94.5)	130 (97.5)	—	0.054
Cholesterol	2177 (88.3)	168.0 (53.1)	—	1840 (74.6)	149.6 (44.5)	—	<0.001	443 (83.9)	123.4 (69.9)	—	404 (76.5)	100.1 (65.9)	—	<0.001
HDL	1951 (79.1)	35.0 (23.0)	—	1817 (73.7)	33.0 (23.8)	—	<0.001	443 (83.9)	33.0 (18.9)	—	397 (75.2)	29.0 (19.9)	—	<0.001
LDL	1952 (79.2)	69.9 (51.5)	1216 (62.3)	1807 (73.3)	54.0 (43.2)	883 (45.2)	<0.001	442 (83.7)	66.5 (45.9)	226 (51.1)	404 (76.5)	49.7 (40.2)	157 (39.9)	<0.001
Non-HDL	2173 (88.1)	113.4 (64.6)	1435 (66.0)	1840 (74.6)	78.6 (58.2)	879 (40.5)	<0.001	443 (83.9)	90.7 (57.3)	222 (50.1)	394 (74.6)	71.6 (51.5)	143 (35.4)	<0.001

Non-HDL calculated using total cholesterol minus HDL cholesterol. LDL calculated using the Friedewald equation. Paired T test used to compare average lab values between all patients at initial versus final encounter* and patients on statins at initial versus final encounter**.

Patients with LDL <70 mg/dL had no difference in amount of HLD, CAD, PAD, HTN, or CVOD (p = 0.854, p = 0.101, p = 0.420, p = 0.657, p = 0.732, respectively) and less amount of DM diagnoses (p = 0.017) than those with LDL >70 mg/dL. Patients with non-HDL <100 mg/dL had no difference in amount of HLD, CAD, DM, HTN, or CVOD (p = 0.2 65, p = 0.604, p = 0.062, p = 0.337, p = 0.869, respectively) but less PAD (p = 0.024) diagnoses than those with non-HDL >100 mg/dL. Patients with LDL <70 mg/dL and non-HDL <100 mg/dL had no difference in amount of HLD, CAD, PAD, DM, HTN, or CVOD diagnoses than those with LDL <70 mg/dL but non-HDL >100 mg/dL (p > 0.050 for all) (Table 4). When comparing LDL calculated by the Friedewald (F-LDL), Sampson (S-LDL), and Martin-Hopkins (M-LDL) equations, the Friedewald equation seemed to be the most liberal with 51.0% of patients F-LDL <70 mg/dL while the Martin-Hopkins equation seemed to be the most conservative with 43.4% of patients M-LDL <70 mg/dL (Table 5). There were no significant differences in diagnoses between F-LDL, S-LDL, and M-LDL (Table 5). When looking at patients who were at an LDL of <70 mg/dL and non-HDL of <100 mg/dL, the Martin-Hopkins and Sampson equations had higher agreement compared to the Friedewald equation (96.8 and 91.5% versus 86.2%, respectively) (Table 6). There were no significant differences in diagnoses between these groups (Table 6).

Table 4.

Diagnosis codes for patients based on non-HDL and LDL values.

	LDL<70	LDL>70	p-value*	Non-HDL<100	Non-HDL>100	p-value**	LDL <70, non-HDL <100	LDL <70, non-HDL >100	p-value***
Number present	922 (51.0%)	885 (49.0%)	—	950 (52.6%)	857 (47.4%)	—	795 (86.2%)	127 (13.8%)	—
HLD	404 (43.8%)	384 (43.4%)	0.854	426 (44.8%)	362 (42.2%)	0.265	357 (44.9%)	47 (37.0%)	0.096
CAD	65 (7.1%)	46 (5.2%)	0.101	61 (6.4%)	50 (5.8%)	0.604	56 (7.0%)	9 (7.0%)	0.986
PAD	175 (19.0%)	155 (17.5%)	0.420	192 (20.2%)	138 (16.1%)	0.024	156 (19.6%)	19 (15.0%)	0.213
DM	144 (15.6%)	104 (11.8%)	0.017	144 (15.2%)	104 (12.1%)	0.062	129 (16.2%)	15 (11.8%)	0.203
HTN	189 (20.5%)	174 (19.7%)	0.657	199 (21.0%)	164 (19.1%)	0.337	171 (21.5%)	18 (14.2%)	0.057
CVOD	34 (3.7%)	30 (3.4%)	0.732	33 (3.5%)	31 (3.6%)	0.869	28 (3.5%)	6 (4.7%)	0.504

Non-HDL calculated using total cholesterol minus HDL cholesterol. LDL calculated using the Friedewald equation. Chi squared analysis used to compare LDL <70 vs LDL >70*, non-HDL <100 vs non-HDL >100**, LDL <70 and non-HDL <100 vs LDL <70, and non-HDL <100***.

Table 5.

Comparison of patient diagnoses between different LDL cholesterol calculations.

	F-LDL<70	F-LDL>70	p-value*	S-LDL<70	S-LDL>70	p-value**	M-LDL<70	M-LDL>70	p-value***
Number present	922 (51.0%)	885 (49.0%)	—	851 (47.1%)	956 (52.9%)	—	785 (43.4%)	1022 (56.6%)	—
HLD	404 (43.8%)	384 (43.4%)	0.854	381 (44.8%)	407 (42.6%)	0.347	3654 (45.1%)	434 (42.5%)	0.264
CAD	65 (7.1%)	46 (5.2%)	0.101	61 (7.2%)	50 (5.2%)	0.087	55 (7.01%)	56 (5.48%)	0.180
PAD	175 (19.0%)	155 (17.5%)	0.420	159 (18.7%)	171 (17.9%)	0.662	150 (19.1%)	180 (17.6%)	0.415
DM	144 (15.6%)	104 (11.8%)	0.017	133 (15.6%)	115 (12.0%)	0.026	124 (15.8%)	124 (12.13%)	0.025
HTN	189 (20.5%)	174 (19.7%)	0.657	178 (20.9%)	363 (19.4%)	0.407	169 (21.5%)	194 (19.0%)	0.181
CVOD	34 (3.7%)	30 (3.4%)	0.732	31 (3.6%)	33 (3.4%)	0.827	29 (3.7%)	35 (3.42%)	0.759

F-LDL calculated using the Friedewald equation. S-LDL calculated using the Sampson equation. M-LDL calculated using the Martin-Hopkins equation. Chi squared analysis used to compare F-LDL < 70 vs F-LDL > 70*; S-LDL < 70 vs S-LDL > 70**; M-LDL < 70 vs M-LDL > 70***.

Table 6.

Comparison of patient diagnoses between different LDL cholesterol calculations and non-HDL values.

	F-LDL <70, non-HDL <100	F-LDL <70, non-HDL >100	p-value*	S-LDL <70, non-HDL <100	S-LDL <70, non-HDL >100	p-value**	M-LDL <70, non-HDL <100	M-LDL <70, non-HDL >100	p-value***
Number present	795 (86.2%)	127 (13.8%)	—	779 (91.5%)	72 (8.5%)	—	760 (96.8%)	25 (3.2%)	—
HLD	357 (44.9%)	47 (37.0%)	0.096	353 (45.3%)	28 (38.9%)	0.294	343 (45.1%)	11 (44.0%)	0.911
CAD	56 (7.0%)	9 (7.0%)	0.986	56 (7.2%)	5 (6.9%)	0.939	54 (7.1%)	1 (4.0%)	0.550
PAD	156 (19.6%)	19 (15.0%)	0.213	149 (19.1%)	10 (13.9%)	0.275	146 (19.2%)	4 (16.0%)	0.688
DM	129 (16.2%)	15 (11.8%)	0.203	126 (16.2%)	7 (9.7%)	0.149	122 (16.1%)	2 (8.0%)	0.277
HTN	171 (21.5%)	18 (14.2%)	0.057	169 (21.7%)	9 (12.5%)	0.066	165 (21.7%)	4 (16.0%)	0.494
CVOD	28 (3.5%)	6 (4.7%)	0.504	28 (3.6%)	3 (4.2%)	0.804	29 (3.8%)	0 (0.0%)	0.320

F-LDL calculated using the Friedewald equation. S-LDL calculated using the Sampson equation. M-LDL calculated using the Martin-Hopkins equation. Non-HDL calculated using total cholesterol minus HDL cholesterol. Chi squared analysis used to compare F-LDL < 70, non-HDL <100 vs F-LDL < 70, non-HDL >100*; S-LDL < 70, non-HDL <100 vs S-LDL < 70, non-HDL >100**; M-LDL < 70, non-HDL <100 vs M-LDL < 70, non-HDL >100***.

Discussion

In the present study, we demonstrate that among vascular surgery patients, approximately two-thirds were not at a treatment target of LDL <70 md/dL or non-HDL of <100 mg/dL at their first encounter. Patients remained undertreated with approximately half of patients not at or below LDL or non-HDL thresholds at their final encounter. Current guidelines for PAD and CVOD recommend patients to be on statin therapy with lipid-lowering targets as a secondary treatment threshold. Our results demonstrate that we may be vastly under-performing proper medical optimization with only about one-fourth of patients on a statin at their final encounter and approximately one-tenth of patients being treated with a high-intensity statin. The present study highlights the fact that vascular surgery patients, who are often at high risk for ASCVD, may be significantly undertreated and are not currently being optimized at current guideline thresholds.

Non-HDL has been proposed as a more reliable measurement than LDL because it accounts for all atherogenic lipoproteins, whereas LDL is indirectly estimated as a function of TG, HDL, and TC, most often via the Friedewald equation.²⁷ Measurement of TG have been plagued by a high degree of variability even in the fasting state.³⁰ Non-HDL can be measured on a non-fasting patient and easily calculated from a lipid profile without the use of TG and thus is less subject to error induced by high TG in a non-fasting state.¹¹ Furthermore, non-HDL is a better predictor of adverse cardiovascular outcomes amongst patients on lipid-lowering therapy than LDL.^14,15 Thus, non-HDL remains a promising adjunct for ASCVD risk stratification and appears to be more reliable and accurate than LDL. In our study, using different LDL equations (Friedewald, Sampson, and Martin-Hopkins) lead to differences in the percentage of patients who were at LDL <70 mg/dL and non-HDL <100 mg/dL, with the Martin-Hopkins equation demonstrating the highest congruence of patients at both LDL and non-HDL goals (96.8%). These results support that non-HDL may be a more reliable measure and demonstrate that the use of the Friedewald equation performs the worst. Given the differences seen between LDL equations, there is potential to optimize medical management beyond current high-intensity statin therapy using non-HDL as a treatment marker. However, further investigation is needed to determine the association of non-HDL cholesterol with disease burden and post-operative outcomes.

When comparing patients who were above and below LDL of 70 mg/dL and those above and below non-HDL of 100 mg/dL, there was no difference in patient diagnoses of CAD, HTN, or CVOD at their final encounter. Interestingly, patients with LDL >70 mg/dL had lower amount of DM diagnoses and patients with non-HDL >100 mg/dL had a lower amount of HLD and PAD diagnoses. These findings were irrespective of LDL calculation (Friedewald, Sampson, and Martin-Hopkins). However, these findings were conducted with a univariate analysis and thus are limited given that we did not account for other factors that could contribute to diagnoses such as CAD, HTN, DM, PAD, and CVOD. The former could possibly be explained by metabolic derangements leading to higher TG, thus erroneously calculating lower LDL by the Friedewald equation. However, our sensitivity analysis did not confirm this to be the case. Although these results seemingly support that lipid-based treatment thresholds do not improve outcomes, our dataset is limited and not appropriate to answer this question. On the contrary, we feel it informs on the high-risk nature of patients being referred to vascular surgery.

In our study, patients who were on statins demonstrated a significant decrease in their non-HDL and LDL when comparing initial and final encounters. Moreover, patients treated with statins had more patients at LDL and non-HDL treatment thresholds. Although this dataset included patients seen by vascular surgery providers, a majority of medical management is oftentimes multi-disciplinary with many patients being managed by cardiology, vascular medicine, and primary care. Regular dose titration to LDL of <70 mg/dL is not routinely done by vascular surgeons; our results support that statin therapy is an effective means of lipid lowering and risk reduction via secondary treatment thresholds. However, our findings highlight that there is significant room for improvement in the medical optimization of vascular surgery patients. Moreover, our results highlight potential under-prescribing of statins to patients with peripheral vascular and cerebrovascular disease given the higher rates of statin usage in patients with CAD when compared to PAD and CVOD. Given the astoundingly low percentage of patients on statins, improved methods of identifying patients not optimally managed medically are warranted. A multi-modal approach including vascular surgeons, cardiovascular medicine, and primary care providers is needed to ensure adequate lipid-lowering treatment of vascular surgery patients. Vascular surgery providers must place increasing emphasis on ensuring their patients are treated appropriately with statin therapy through improved communication with primary care providers and increased prescribing of statin therapy when indicated.

Interestingly, there was decrease in HDL between first and final encounters for all patients and those on statins, which could be explained via a few mechanisms which were not captured in our database. Given that inactivity has been demonstrated to be suggested to be a cause of lower HDL, vascular surgery patients with progressive disease and those undergoing surgical interventions may have become less active over time thus contributing to a decrease in HDL.^31,32 Additionally, numerous cardiovascular medications, antipsychotics, anticonvulsants, hormones, and certain immunosuppressives have been demonstrated to have a negative impact on lipid profiles.³³ Some medications demonstrate a class effect while others are dose dependent and drug specific.³³ Nonetheless, these were not accounted for in our dataset.

This study was a single-center retrospective study and thus has limitations that must be kept in mind. Although the U-M has a large patient catchment area, the pre-dominant vascular surgery patients are non-Hispanic white males which was reflected in our cohort. Vascular disease affects all races and genders, and thus, our results may not be completely generalizable to the entire population. Furthermore, our study depicts only medication trends from a single center rather than nationwide. Given that LDL and cholesterol treatment guidelines have changed over time with previous recommendations of LDL values <100 mg/dL, patients who were seen and managed earlier in the cohort may have been at their treatment goal based on recommendations at that time.⁸ Our study relied on ICD9/ICD10 codes which may vary between institutions nationally and hospital coders at a single institution and do not include quantifiable or robust information about disease burden. Retrospective studies are subject to limitations associated with data from the electronic health record including missing data, such as laboratory values performed at other institutions. If a diagnosis and/or medication was not associated with an encounter, then it would not have been captured in our study and thus could impact our results. Additionally, the diagnoses evaluated in this study are not equivalent to each other and are not indications alone for statin usage, but rather strong risk factors for poor outcomes and a major component of risk estimation equations. Diagnoses evaluated in this study are chronic and thus would likely not change between patient encounters. Furthermore, our study did not assess for other lipid-modifying treatment modalities such as lifestyle modifications, PCSK9 inhibitors, bile acid sequestrants, or fibrates. Given that our dataset consisted of outpatient and inpatient encounters, we cannot ensure that all lipid panels were obtained in the fasting state which could have affected that accuracy of our results. Future studies are needed to elucidate the association of non-HDL levels on disease burden and patient outcomes.

Conclusion

In our study, only one-fourth of statin-eligible vascular surgery patients were treated with statin therapy. Furthermore, approximately half of patients seen by vascular surgery were not at secondary treatment threshold with elevated LDL and non-HDL at their initial and final encounters. These results highlight that vascular surgery patients may be undertreated and not medically optimized. There is significant room for improvement in medical therapy for vascular surgery patients. A multi-disciplinary approach from vascular surgery, cardiovascular, and general medical providers is needed to increase statin usage in patients with PAD and/or CVOD. Further research is needed to identify provider and patient barriers for statin usage and optimal medical therapy. Non-HDL is a promising adjunct for ASCVD risk stratification and serves as a secondary treatment target for further medical optimization. However, future studies are needed to explore the association of non-HDL cholesterol with atherosclerotic disease burden and post-operative outcomes among vascular surgery patients.

Supplemental Material

Supplemental Material - Non-High density lipoprotein cholesterol and treatment targets in vascular surgery patients

Supplemental Material for Non-High density lipoprotein cholesterol and treatment targets in vascular surgery patients by Drew J Braet, Kian Pourak, Vibav Mouli, Itai Palmon, Dan Dinh, Nicholas H Osborne, Chandu Vemuri and Eric J Brandt in Vascular

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Drew J Braet

Supplemental Material

Supplemental material for this article is available online.

References

Benjamin

Blaha

Chiuve

, et al. Heart Disease and Stroke Statistics—2017 Update: A Report From the American Heart Association. Circulation 2017; 135: e146–e603. DOI: 10.1161/CIR.0000000000000485

Libby

Buring

Badimon

, et al. Atheroscler. Nat Rev Dis Primers 2019; 5: 56. DOI: 10.1038/s41572-019-0106-z

American Diabetes Association . Cardiovascular Disease and Risk Management. Diabetes Care 2016; 39: S60–S71. DOI: 10.2337/dc16-S011

Gerhard-Herman

Gornik

Barrett

, et al. AHA/ACC Guideline on the Management of Patients With Lower Extremity Peripheral Artery Disease: Executive Summary: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2016; 135: e686–e725. DOI: 10.1161/CIR.0000000000000470

Stone

Robinson

Lichtenstein

, et al. ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults. JACC 2013; 63: 2889–2934.

Catapano

Graham

De Backer

, et al. ESC/EAS Guidelines for the Management of Dyslipidaemias. Eur Heart J 2016; 37: 2999–3058.

Jacobson

Ito

Maki

, et al. National Lipid Association Recommendations for Patient-Centered Management of Dyslipidemia: Part 1—Full Report. J Clin Lipiol 2015; 9: 129–169.

National Cholesterol Education Program (NCEP) . Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 2002; 106(25): 3143–3421.

Arnett

Blumenthal

Albert

, et al. ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2019; 140: e596–e646. DOI: 10.1161/CIR.0000000000000678

10.

United States Preventative Services Task Force (USPSTF) . Statin Use for the Primary Prevention of Cardiovascular Disease in Adults: Preventive Medication. USPSTF, 2016.

11.

Carr

Hooper

Sullivan

, et al. Non-HDL-cholesterol and apolipoprotein B compared with LDL-cholesterol in atherosclerotic cardiovascular disease risk assessment. Pathology 2019; 51: 148–154.

12.

The Emerging Risk Factors Collaboration*. Major Lipids, Apolipoproteins, and Risk of Vascular Disease. JAMA 2009; 302: 1993–2000.

13.

Sandesara

Virani

Fazio

, et al. The forgotten lipids: triglycerides, remnant cholesterol, and atherosclerotic cardiovascular disease risk. Endocr Rev 2019; 40: 537–557.

14.

Brunner

Waldeyer

Ojeda

, et al. Application of non-HDL cholesterol for population-based cardiovascular risk stratification: results from the Multinational Cardiovascular Risk Consortium. Lancet 2019; 394: 2173–2183.

15.

Kong

Peng

. Evidence for changing lipid management strategy to focus on non-high density lipoprotein cholesterol. Lipids Health Dis 2019; 18: 134.

16.

Puri

Nissen

Shao

, et al. Non-HDL Cholesterol and Triglycerides: Implications for Coronary Atheroma Progression and Clinical Events. ATVB 2016; 36: 2220–2228.

17.

Kastelein

JJP

van der Steeg

Holme

, et al. Lipids, Apolipoproteins, and Their Ratios in Relation to Cardiovascular Events With Statin Treatment. Circulation 2008; 117: 3002–3009.

18.

Sniderman

Williams

Contois

, et al. A Meta-Analysis of Low-Density Lipoprotein Cholesterol, Non-High-Density Lipoprotein Cholesterol, and Apolipoprotein B as Markers of Cardiovascular Risk. Circ Cardiovasc Qual Outcomes 2011; 4: 337–345.

19.

Bertrand

Saulnier

Potier

, et al. Plasma concentrations of lipoproteins and risk of lower-limb peripheral artery disease in people with type 2 diabetes: the SURDIAGENE study. Diabetologia 2021; 64: 668–680.

20.

Liu

Zhang

Xia

, et al. Relationship between the non-HDLc-to-HDLc ratio and carotid plaques in a high stroke risk population: a cross-sectional study in China. Lipids Health Dis 2020; 19: 168.

21.

Wang

Kang

, et al. Association between cumulative exposure to different lipid parameters and risk of newly developed carotid plaque. Stroke Vasc Neurol 2021; 6: 359–365.

22.

Ding

Chen

Shi

, et al. Association between nontraditional lipid profiles and peripheral arterial disease in Chinese adults with hypertension. Lipids Health Dis 2020; 19: 231.

23.

Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001; 285: 2486–2497.

24.

AbuRahma

Avgerinos

Chang

, et al. Society for Vascular Surgery clinical practice guidelines for management of extracranial cerebrovascular disease. JVS 2022; 75: 4S–22S.

25.

Aboyans

Ricco

Bartelink

MLEL

, et al. ESC Guidelines on the Diagnosis and Treatment of Peripheral Arterial Diseases, in collaboration with the European Society for Vascular Surgery (ESVS). Eur Heart J 2017; 39: 763–816.

26.

Lee

Shin

, et al. Attainment of low-density lipoprotein cholesterol goal after endovascular treatment is associated with reduced cardiovascular events in patients with peripheral arterial disease. JVS 2016; 63: 756–763.

27.

Friedewald

Levy

Fredrickson

. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972; 18: 449–502.

28.

Sampson

Ling

Sun

, et al. A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia. JAMA Cardiol 2020; 5(5): 540.

29.

Martin

Blaha

Elshazly

, et al. Friedewald-plestimated versus directly measured low-density lipoprotein cholesterol and treatment imications. JACC 2013; 62(8): 732–739.

30.

Smith

Cooper

Myers

, et al. Biological variability in concentrations of serum lipids: sources of variation among results from published studies and composite predicted values. Clin Chem 1993; 39: 1012–1022.

31.

O’Donovan

Stensel

Hamer

, et al. The association between leisure-time physical activity, low HDL-cholesterol and mortality in a pooled analysis of nine population-based cohorts. Eur J Epidemiol 2017; 32(7): 559–566.

32.

Crichton

Alkerwi

. Physical activity, sedentary behavior time and lipid levels in the Observation of Cardiovascular Risk Factors in Luxembourg study. Lipids Health Dis 2015; 14: 87.

33.

Herink

Ito

. Medication Induced Changes in Lipid and Lipoproteins. Endotext, 2000, https://www-ncbi-nlm-nih-gov.proxy.lib.umich.edu/books/NBK326739 (accessed 20 June 2022).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB

Non-high-density lipoprotein cholesterol and treatment targets in vascular surgery patients

Abstract

Introduction

Methods

Results

Discussion

Keywords

Introduction

Methods

Study population

Data analysis

Results

Discussion

Conclusion

Supplemental Material

Supplemental Material - Non-High density lipoprotein cholesterol and treatment targets in vascular surgery patients

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

Supplemental Material

References

Supplementary Material