- Email: [email protected]
Prevalence of familial hypercholesterolemia in patients with acute coronary syndrome in Japan: Results of the EXPLORE-J study
Atherosclerosis 277 (2018) 362e368
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Prevalence of familial hypercholesterolemia in patients with acute coronary syndrome in Japan: Results of the EXPLORE-J study Mariko Harada-Shiba a, *, Junya Ako b, Hidenori Arai c, Atsushi Hirayama d, Yoshitaka Murakami e, Atsushi Nohara f, Asuka Ozaki g, Kiyoko Uno g, Masato Nakamura h a Department of Molecular Innovation in Lipidology, National Cerebral and Cardiovascular Center Research Institute, 5-7-1 Fujishiro-dai, Suita, Osaka, 5658565, Japan b Department of Cardiovascular Medicine, Kitasato University, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa, 252-0375, Japan c Department of Geriatric Medicine, National Center for Geriatrics and Gerontology, 7-430 Morioka-cho, Obu City, Aichi, 474-8511, Japan d Division of Cardiology, Nihon University School of Medicine, 30-1 Oyaguchi Kamicho, Itabashi-ku, Tokyo, 173-8610, Japan e Department of Medical Statistics, Toho University, 5-21-16 Omori-Nishi, Ota-ku, Tokyo, 143-8540, Japan f Department of Cardiology, Kanazawa University of Graduate School of Medical Sciences, 13-1 Takara-machi, Kanazawa, Ishikawa, 920-8641, Japan g Sanofi, Tokyo Opera City Tower, 3-20-2 Nishi Shinjuku, Tokyo, 163-1488, Japan h Division of Cardiovascular Medicine, Toho University Ohashi Medical Center, 2-17-6 Ohashi Meguro, Tokyo, 153-8515, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 March 2018 Received in revised form 30 May 2018 Accepted 14 June 2018
Background and aims: Prevalence of familial hypercholesterolemia (FH), a common genetic disorder with a high risk for coronary artery disease (CAD), is high among CAD patients; however, data on FH prevalence among acute coronary syndrome (ACS) patients are limited. EXPLORE-J is the largest registry to diagnose FH among Japanese ACS patients using the 2012 Japan Atherosclerosis Society guidelines. Methods: This prospective study consecutively recruited patients between April 2015 and August 2016 at 59 sites. Low-density lipoprotein cholesterol (LDL-C) levels, family history of premature CAD, presence of tendon xanthomas, and Achilles tendon radiograms were recorded at baseline. The prevalence rate of FH in patients with ACS was estimated with 95% CI. Results: Of 1944 analyzed patients (mean age, 66.0 years; men, 80.3%), 52 (2.7% [95% CI: 2.0e3.5]) had FH. Thirty-one (1.6%) had LDL-C 180 mg/dL and Achilles tendon thickness (ATT) 9 mm, 8 (0.4%) had LDL-C 180 mg/dL and family history of premature CAD, 10 (0.5%) had ATT 9 mm and family history of premature CAD, and 3 (0.2%) met all the criteria. FH patients were younger than those without FH (59.5 [12.5] vs. 66.2 [12.1] years; p < 0.001). More patients with premature ACS (men, <55 years; women, <65 years) than without (4.7% [95% CI: 2.9e7.2] vs. 2.1% [1.4e3.0]) had FH. Conclusions: FH prevalence is at least five-fold higher in ACS patients than in the general population, especially in patients with premature ACS onset and ATT 9 mm. FH screening in ACS patients is therefore clinically important and critical. © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Familial hypercholesterolemia Achilles tendon thickness Acute coronary syndrome Coronary artery disease Low-density lipoprotein
1. Introduction Familial hypercholesterolemia (FH) is an inherited, autosomal
* Corresponding author. Department of Molecular Innovation in Lipidology, National Cerebral and Cardiovascular Center Research Institute, 5-7-1 Fujishiro-dai, Suita, Osaka, 565-8565, Japan. E-mail addresses: [email protected] (M. Harada-Shiba), [email protected] (J. Ako), [email protected] (H. Arai), [email protected] (A. Hirayama), [email protected] (Y. Murakami), [email protected] (A. Nohara), Asuka.Ozaki@sanofi.com (A. Ozaki), Kiyoko.Uno@sanofi.com (K. Uno), [email protected] (M. Nakamura).
dominant disease caused by abnormalities in the genes that code for low-density lipoprotein (LDL) receptors and related molecules. It is characterized by hyper-LDL cholesterolemia, early-onset coronary artery disease (CAD), and tendon or cutaneous xanthomas [1e3]. Globally, the estimated prevalence of FH is approximately 1 in 500 [3], although recent reports have estimated a prevalence of 1 in 250 [4], and even as high as 1 in 223 in a study in a Danish population [5]. The estimated number of cases in Japan is 300,000 [3], although a study conducted in the Hokuriku district of Japan reported a prevalence as high as 1 in 208 [6]. However, less than 1% of people affected by FH in Japan are estimated to receive a diagnosis;
https://doi.org/10.1016/j.atherosclerosis.2018.06.856 0021-9150/© 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
this problem of underdiagnosis may stem from the fact that patients with FH can be easily overlooked among the vast majority of individuals with cardiovascular disease caused by more common risk factors and by lack of knowledge among clinicians [7]. A survey conducted among primary care physicians worldwide showed that awareness of FH guidelines was low in Western countries such as the UK (61%), but was even lower in Japan (47%); this may possibly explain the underdiagnoses of FH in Japan [8]. The 2012 Japan Atherosclerosis Society (JAS) guidelines were more relaxed than most global guidelines and aimed for an LDL cholesterol (LDL-C) level of less than 100 mg/dL (2.6 mmol/L) or a 50% decrease in pretreatment LDL-C levels in patients with FH for primary prevention because of insufficient evidence to demonstrate that lower levels of LDL-C are required to reduce the risk of CAD in Japanese patients [9]. In the absence of stringent guidelines for LDL-C reduction in Japan and increased westernization of the Japanese lifestyle, the risk for CAD faced by patients with FH is high, and an efficient and accurate set of diagnostic criteria for FH are needed for early diagnosis and mitigation of risk for cardiovascular disease. According to the 2012 JAS guidelines for the diagnosis of FH, an adult (15 years) meeting two or more of the following criteria is diagnosed with FH: hyper-LDL cholesterolemia (untreated LDL-C level 180 mg/dL [4.7 mmol/L]); tendon xanthomas (including Achilles tendon hypertrophy [defined as Achilles tendon thickness (ATT) 9 mm]) or xanthoma tuberosum; and family history of FH or premature CAD (<55 years in men and <65 years in women) within first- and second-degree relatives. In addition, the diagnosis of FH should be made after eliminating the possibility of secondary hyperlipidemia [3]. It is important to note that, unlike any other set of criteria in any other country, JAS guidelines integrate the measurement of ATT, an easily accessible and interpretable modality, into the diagnosis of FH. In many patients, the initial clinical manifestation of FH is acute coronary syndrome (ACS) [10] that usually occurs at a younger age due to exposure to very high LDL-C levels from birth [11]. Therefore, the proportion of patients with FH is expected to be greater in patients with ACS than in the general population. A prospective, observational cohort study in patients hospitalized for ACS in Switzerland reported that according to the Dutch Lipid Clinic Network algorithm, 1.6% (95% confidence interval [CI]: 1.3e2.0%) had probable/definite FH and 17.8% (16.8e18.9%) had possible FH, whereas according to the Simon Broome Register algorithm, 5.4% (4.8e6.1%) had possible FH [12]. Ohmura et al. reported the results of a prospective, multicenter registry study to estimate the prevalence of FH in patients with ACS. Overall, 5.7% of patients with ACS were diagnosed with FH. Of those patients with ATT 9 mm, approximately 28.6% had FH [13]. However, because this study was limited by its small sample size, these results required confirmation in a larger population. EXPLORE-J is a prospective, large-scale, observational study of patients presenting with ACS conducted at 59 sites in Japan. To date, it is the largest registry study in Japan to focus on the high-risk ACS population and use ATT data for diagnosis of FH [14]. The objective of our current analysis was to describe the prevalence of FH in Japanese patients with ACS using the 2012 JAS guidelines and data from the EXPLORE-J registry study. 2. Patients and methods 2.1. Study design In this prospective, observational study conducted at 59 sites across Japan, patients who presented with ACS were consecutively enrolled between April 2015 and August 2016. Details of the study
363
design have been previously reported [14]. Briefly, patients were registered within 7 days following hospitalization for ACS. Data were collected at visits 1 through 5 during the 2-year observation period using an electronic case report form (Fig. 1). This study was conducted in compliance with the Declaration of Helsinki (amended in October 2013) and the Ethical Guidelines for Medical and Health Research Involving Human Subjects (enacted on 22 December 2014), and was approved by an ethical review committee at each site. All patients provided written informed consent prior to participation. 2.2. Patients Detailed inclusion and exclusion criteria have been previously reported [14]. Briefly, patients aged 20 years and over who present with ACS, including ST elevation myocardial infarction (STEMI), noneST elevation myocardial infarction (NSTEMI), or unstable angina, at the time of informed consent were eligible for inclusion. Patients with chest pain caused by prespecified comorbidities or instent thrombosis, patients enrolled in other interventional studies that could affect the lipid profile, and patients with conditions that could affect the monitoring and the natural course of the disease condition after treatment were excluded. 2.3. Assessments The highest available measurement of LDL-C obtained at or before visit 1 was used for analysis of FH. The LDL-C measurement (regardless of method) prior to hospitalization without treatment and the first measurement after hospitalization were reported. LDL-C levels were checked at visit 1 (within 14 days of hospitalization for ACS) and at every subsequent visit over the 2-year observation schedule (including both direct and calculated measurements). As previously described [15], ATT was measured at each site and at the central reading laboratory by three highly trained readers who were blinded to the patients' clinical characteristics using radiographs taken during hospitalization for registration in the study; however, radiographic measurement performed by visit 3 was also acceptable. Information on family history of CAD, ischemic cerebral infarction, and hypercholesterolemia within first- and second-degree relatives was collected at enrollment. Only family history of premature CAD was used for FH diagnosis. 2.4. Statistical analysis Details of the statistical analysis in this study have been previously reported [14]. Sample size was calculated to assess the persistent cardiovascular risk (major adverse cardiac events [MACE]) throughout the 2-year observation period, beginning with the index event. Based on the PACIFIC registry, in which the incidence of MACE was 6.4% at 2 years, a sample size of 2000 patients yielded a precision of ±1% in the incidence of MACE with a 95% CI of 0.053e0.074. Demographic variables were described by mean, median, standard deviation (SD), and range for continuous data, and by proportion in each category for categorical data. Fisher's exact test or Mann-Whitney U test was used for comparison of categorical variables according to nominal or ordinal scale. Student's t-test or Wilcoxon rank sum test was used for continuous variables as appropriate. A two-sided p value of <0.05 was considered significant. The prevalence of FH in patients with ACS was estimated with the associated 95% CI.
364
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
Fig. 1. Study design. a Patients were registered within 7 days of ACS. bRadiographs of the Achilles tendon will be obtained during hospitalization for registration as a rule, but a radiograph obtained by visit 3 is acceptable. ACS: acute coronary syndrome; LDL-C: low-density lipoprotein cholesterol; V: visit.
3. Results 3.1. Demographics and baseline characteristics Overall, 2016 patients were registered; of these, 1944 were included in the analysis (reasons for exclusion from the analysis population: failure to obtain informed consent within 7 days of hospitalization, 62; unapproved informed consent, 4; no informed consent obtained, 2; duplicate entry, 2; erroneous entry, 1; withdrawal due to achievement of target sample size, 1). The mean (SD) age was 66.0 (12.2) years, and the mean (SD) body mass index (BMI) was 24.2 (3.6) kg/m2; 80.3% of patients were men (Table 1). Cardiovascular risk factors such as hypertension (73.4%), dyslipidemia (77.8%), and diabetes (34.9%) were common. Overall, 625 (32.2%) patients were on lipid-modifying therapy (LMT): statins (530 [27.3%]), intensive statins (30 [5.7%]), eicosapentaenoic acid/docosahexaenoic acid (70 [3.6%]), ezetimibe (40 [2.1%]), and fibrates (35 [1.8%]).
for xanthomas, but family history of premature CAD was missing). On exclusion of patients with any missing data (n ¼ 1579), the diagnosis rate for FH was 3.1% (95% CI: 2.3e4.1). 3.3. Comparison of background data/baseline characteristics Baseline characteristics except age were similar between patients diagnosed with FH and those not diagnosed with FH. The mean (SD) age was 59.5 (12.5) years in patients diagnosed with FH versus 66.2 (12.1) years in patients without FH (p < 0.001) (Table 1). Proprotein convertase subtilisin/kexin 9 levels were comparable between patients diagnosed with FH and without FH. A lower proportion of patients with FH diagnosis than those without FH previously used LMTs (17.3% vs. 32.6%, p ¼ 0.023). However, among patients who had previously used statins, none of the seven patients diagnosed with FH and 30/523 (5.7%) without a diagnosis of FH were on intensive statins. 3.4. Subgroup analysis by particular diagnostic criterion for FH
3.2. Diagnosis of FH Of the 1944 analyzed patients, 52 (2.7% [95% CI: 2.0e3.5]) met two or more of the JAS criteria and were diagnosed with FH (Fig. 2). Most patients diagnosed with FH met only two criteria (49/1944 [2.5%]). Of these, 31/1944 (1.6%) had LDL-C 180 mg/dL and ATT 9 mm; 8/1944 (0.4%) had LDL-C 180 mg/dL and family history of premature CAD; and 10/1944 (0.5%) had ATT 9 mm and family history of premature CAD. Only 3/1944 (0.2%) patients met all three diagnostic criteria for FH. Of note, 19.3% of the patients met one of the three criteria. Of the analyzed patients, FH was indeterminable for 48 patients due to missing data (e.g., positive for LDL-C 180 mg/dL, negative
Overall, 23.2% (42/181) of patients who had LDL-C 180 mg/dL, 29.1% (44/151) of those who had xanthomas, and 13.9% (21/151) of those with family history of premature CAD were diagnosed with FH (Table 2). Data on family history of premature CAD were missing in 355 patients (18.3%), and this was the most common missing information. 3.5. Subgroup analysis by baseline characteristics A subgroup analysis of patients with FH according to baseline characteristics was conducted (Table 3). The proportion of patients with premature ACS onset (men, <55 years; women, <65 years)
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
365
Table 1 Demographics and baseline characteristics. All patients
Age (years), mean (SD) Men, n (%) BMI (kg/m2), mean (SD) PCSK9ad (ng/mL), median (Min, Max) LDL-C (mg/dL), mean (SD) Highest available level at baseline (1, 2, 3, or 4) 1) Measurement without medication prior to hospitalization 2) First measurement after hospitalization 3) Direct method 4) Calculated ACS type, n (%) STEMI NSTEMI Unstable angina Medical history, n (%) Coronary artery diseasew Cerebrovascular accidentw Peripheral artery diseasew Diabetes mellitus Hypertension Dyslipidemia Therapy before hospitalization Any LMT, n (%) Statin Intensive statiny,z Fibrates EPA/DHA Ezetimibe Antiglycemic (except insulin) Antiglycemic (insulin)
All patients (n ¼ 1944)
FH diagnosis (n ¼ 52)
No FH diagnosis (n ¼ 1892)
p value
66.0 (12.2) 1561 (80.3) 24.2 (3.6)c 356.0 (103.0, 1158.0)e
59.5 (12.5) 41 (78.8) 25.1 (3.6) 373.0 (165.0, 810.0)f
66.2 (12.1) 1520 (80.3) 24.2 (3.6)d 354.0 (103.0, 1158.0)g
<0.001a 0.727b 0.066a 0.078h
127.2 (39.5)i 132.6 (38.2)l 121.3 (40.0)o 101.8 (33.2)q 99.4 (31.9)t
201.2 187.1 191.9 153.8 144.5
125.1 (37.0)j 130.5 (36.8)n 119.2 (37.3)p 100.4 (31.3)s 98.2 (30.1)v
<0.001k <0.001a <0.001k <0.001k <0.001k
1195 (61.5) 309 (15.9) 440 (22.6)
31 (59.6) 8 (15.4) 13 (25.0)
1164 (61.5) 301 (15.9) 427 (22.6)
0.901b
355 (18.3) 149 (7.7) 37 (1.9) 679 (34.9) 1427 (73.4) 1512 (77.8)
5 (9.6) 3 (5.8) 1 (1.9) 21 (40.4) 33 (63.5) 50 (96.2)
350 (18.5) 146 (7.7) 36 (1.9) 658 (34.8) 1394 (73.7) 1462 (77.3)
0.060x 0.571x 0.801x 0.461b 0.112b <0.001b
625 (32.2) 530 (27.3) 30 (5.7)aa 35 (1.8) 70 (3.6) 40 (2.1) 382 (19.7) 77 (4.0)
9 7 0 1 1 2 7 1
616 (32.6) 523 (27.6) 30 (5.7)ac 34 (1.8) 69 (3.6) 38 (2.0) 375 (19.8) 76 (4.0)
0.023b 0.026b 1.000b 0.616b 1.000b 0.290b 0.293b 0.721b
(52.9) (36.1)m (58.9) (53.8)r (54.1)u
(17.3) (13.5) (0)ab (1.9) (1.9) (3.8) (13.5) (1.9)
ACS: acute coronary syndrome; BMI: body mass index; EPA/DHA: eicosapentaenoic acid/docosahexaenoic acid; FH: familial hypercholesterolemia; LDL-C: low-density lipoprotein cholesterol; LMT: lipid-modifying therapy; NSTEMI: noneST elevation myocardial infarction; PCSK9: proprotein convertase subtilisin/kexin 9; SD: standard deviation; STEMI: ST elevation myocardial infarction. a Student's t-test. b Fisher's exact test. c n ¼ 1937. d n ¼ 1885. e n ¼ 1876. f n ¼ 51. g n ¼ 1825. h Wilcoxon rank sum test. i n ¼ 1924. j n ¼ 1872. k Welch's t-test. l n ¼ 380. m n ¼ 14. n n ¼ 366. o n ¼ 1827. p n ¼ 1775. q n ¼ 1439. r n ¼ 39. s n ¼ 1400. t n ¼ 1797. u n ¼ 48. v n ¼ 1749. w Data were taken as Yes, No, or Possible; data for Yes are used in this table. x Mann-Whitney U test. y In patients previously treated with statins. z Intensive statins ¼ atorvastatin 20 mg, rosuvastatin 10 mg, and pitavastatin 4 mg. aa n ¼ 530. ab n ¼ 7. ac n ¼ 523. ad Heterodimer.
diagnosed with FH was higher than the proportion of patients without premature ACS onset and an FH diagnosis (4.7% [95% CI: 2.9e7.2] vs. 2.1% [95% CI: 1.4e3.0]). The prevalence of FH was 8.3% (95% CI: 1.8e22.5) in patients younger than 40 years and 2.6% (95% CI: 1.9e3.4) in those aged 40 years and older. FH was more
frequently diagnosed in patients without prior use of statins than in patients who had previously used statins (3.2% [95% CI: 2.3e4.2] vs. 1.3% [95% CI: 0.5e2.7]). There was no difference in FH prevalence between gender or ACS type.
366
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
Fig. 2. Diagnosis of FH using central review data of ATT. a Tendon xanthomas or ATT 9 mm. ATT: Achilles tendon thickness; CAD: coronary artery disease; FH: familial hypercholesterolemia; LDL-C: low-density lipoprotein cholesterol.
Table 2 Subgroup analysis of patients meeting particular diagnostic criterion for FH. Criteria
All patients All
LDL-C 180 mg/dL <180 mg/dL No data Tendon xanthomas Yes No Physical exam, Yes Physical exam, No ATT 9 mm (central reading) ATT <9 mm (central reading) Family history of premature CAD Yes No No data
Table 3 Subgroup analysis of prevalence rate by baseline characteristics.
FH diagnosis
All patients No FH diagnosis
181 1743 20
42 (23.2) 10 (0.6) 0 (0)
139 (76.8) 1733 (99.4) 20 (100.0)
151 1793 47 1897 116 1624
44 (29.1) 8 (0.4) 13 (27.7) 39 (2.1) 37 (31.9) 15 (0.9)
107 (70.9) 1785 (99.6) 34 (72.3) 1858 (97.9) 79 (68.1) 1609 (99.1)
151 1438 355
21 (13.9) 28 (1.9) 3 (0.8)
130 (86.1) 1410 (98.1) 352 (99.2)
All data are presented as n (%). ATT: Achilles tendon thickness; CAD: coronary artery disease; FH: familial hypercholesterolemia; LDL-C: low-density lipoprotein cholesterol.
4. Discussion To our knowledge, EXPLORE-J is the largest registry study conducted in Japan to evaluate the prevalence of FH using ATT measurement in patients with ACS. We observed that 2.7% of all patients with ACS included in the study met the JAS diagnostic criteria for FH. This rate of FH diagnosis among patients with ACS was at least five times higher than in the general population (0.2e0.5%) [3,6]. While the prevalence of FH observed in this analysis was lower than a previous study in Japan that reported a 5.7% prevalence of FH in patients with ACS [13], it was higher than a Swiss study that reported 1.6% using the Dutch Lipid Clinic Network algorithm [12]. The difference in the rate of diagnosis may have been affected by differences in the diagnostic criteria and nature of datasets. In this study, patients with FH were younger than those without FH (59.5 vs. 66.2 years; p < 0.001). Approximately 5% of patients with premature ACS onset were diagnosed with FH; this finding was comparable to a previous Swiss study wherein 4.8% of patients
All patients Sex Men Women ACS onset Premature Not premature <40 years 40 years ACS onset (men) Premature (<55 years) Not premature (55 years) ACS onset (women) Premature (<65 years) Not premature (65 years) ACS type STEMI NSTEMI Unstable angina Previous use of statins Yes No
All
FH diagnosis
Prevalence [95% CI]
1944
52
2.7 [2.0e3.5]
1561 383
41 11
2.6 [1.9e3.5] 2.9 [1.4e5.1]
426 1518 36 1908
20 32 3 49
4.7 2.1 8.3 2.6
340 1221
16 25
4.7 [2.7e7.5] 2.0 [1.3e3.0]
86 297
4 7
4.7 [1.3e11.5] 2.4 [1.0e4.8]
1195 309 440
31 8 13
2.6 [1.8e3.7] 2.6 [1.1e5.0] 3.0 [1.6e5.0]
530 1414
7 45
1.3 [0.5e2.7] 3.2 [2.3e4.2]
[2.9e7.2] [1.4e3.0] [1.8e22.5] [1.9e3.4]
All data are presented as n. ACS: acute coronary syndrome; CI: confidence interval; FH: familial hypercholesterolemia; NSTEMI: noneST elevation myocardial infarction; STEMI: ST elevation myocardial infarction.
with premature ACS onset also had FH [12], as well as a previous Japanese study in which 7.8% of patients with ACS under 60 years of age had FH [13]. In addition, the prevalence of FH diagnosis seems higher in patients under 40 years of age than in those 40 years and over (8.3% vs. 2.6%). This finding agrees with a previous study conducted in Japan [13], supporting the trend that FH is more common among patients with early onset of ACS, and this subpopulation should be one of the target patient groups for screening of FH. Although the JAS criteria have enabled more convenient diagnosis of FH, there are certain challenges associated with them. As previously mentioned, less than 1% of patients with FH in Japan are
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
estimated to receive an accurate diagnosis of FH [7]. A large proportion of patients with ACS use statins and other LMTs; in this study, 32.2% and 27.3% of patients had a history of use of any LMT and statin, respectively. While beneficial, LMTs may mask “true” LDL-C levels and tendon xanthomas in patients with FH and prevent accurate diagnosis. In addition, ATT is known to have a significant correlation with age [15]. Therefore, ATT in younger patients with FH may not be thick enough to meet the criteria, resulting in a missed diagnosis in such patients. It is important to note that young patients presenting with ACS are a warning sign for FH, and recording their family history of FH and premature CAD is extremely important for diagnosis. As mentioned previously, measurement of ATT for FH diagnosis is unique to this registry. Overall, 7.8% of the patients (151/1944) had tendon xanthomas, and among them, approximately one-third were diagnosed with FH (44 [29.1%]). This finding is similar to that of a previous study conducted in Japan, which reported that 28.3% of patients with ATT 9 mm also had FH [13]. There are two possible reasons for only one-third of patients with ATT 9 mm being diagnosed with FH. First, we did not have untreated LDL-C levels for all patients, and assuming that untreated LDL-C levels were higher than treated LDL-C levels, it is possible that patients with FH may have been misdiagnosed as not having FH because two of the three criteria for FH diagnosis were not met. Second, the criterion regarding FH diagnosis in patients with ATT 9 mm in the JAS guidelines is based on the study by Mabuchi et al. [15]; however, the patients in that study may not have used statins or other LMTs, unlike the patients in the current study. This may possibly have influenced the threshold of ATT 9 mm for FH diagnosis. Further studies may be needed to help us understand the appropriateness of using ATT as a diagnostic criterion for FH diagnosis in the general population or certain populations such as younger patients or patients with ACS. It has also been previously reported that LDL-C levels significantly decrease after an episode of acute myocardial infarction; LDL-C decreases by 31% on the day after the episode to as much as 48% on day 7 after the episode [16]. This means that diagnosis of FH using LDL-C levels just after an acute myocardial infarction could lead to underdiagnosis. Even though the benefits of LMT to lower LDL-C levels in FH patients are well known, reduced use of lipid-lowering treatment may be observed in Japan because until recently in the Japanese guidelines, a lower LDL-C target of less than 100 mg/dL (2.6 mmol/ L) was recommended [9]. In the current study, fewer patients diagnosed with FH reported previous use of statins than those without FH (13.5% vs. 27.6%). It may be inferred that patients who had not used statins previously had higher LDL-C and therefore met the diagnosis criteria for FH. Alternatively, it may be inferred that lower LDL-C levels in patients previously treated with statins may have possibly led to underestimation of FH prevalence. Nevertheless, patients with FH have been reported to have a 20-fold higher risk of CAD without treatment [17]; however, FH is treatable, and the risk of CAD can be reduced after a successful diagnosis. This requires aggressive lipid lowering to achieve LDL-C reduction of at least 50% or more [17]; therefore, an effort to ensure the timely diagnosis of FH followed by an aggressive treatment regimen is of the utmost importance for patients with FH in Japan. Some of the limitations associated with this study were the limited availability of data on untreated LDL-C levels and lack of data on family history of FH. Overall, untreated LDL-C data were only available in selected patients, including 380 patients with untreated LDL-C levels prior to hospitalization. As per the JAS guidelines, an untreated LDL-C level 180 mg/dL is a criterion for FH; therefore, assessment of untreated and treated LDL-C levels together could lead to an underestimation of FH prevalence. However, note that in patients for whom untreated LDL-C levels
367
prior to hospitalization were unavailable, the highest available measurement of LDL-C obtained at baseline, including the first measurement after hospitalization and the measurement at visit 1, was used for analysis of FH in this study. We believe that this may have helped to minimize underestimation of FH prevalence to some extent. Family history was particularly difficult to collect, and 355 patients had missing data after repeated efforts to collect data. With a numerically higher rate of FH prevalence when analysis was done for patients with data for all three diagnostic criteria available (3.1% vs. 2.7%), it is possible that the availability of all data would have provided us with a more accurate and potentially higher estimate of FH prevalence. Additionally, registry data are considered inherently susceptible to bias [18]. However, we avoided selection bias by enrolling consecutive patients. Finally, this study was restricted to the Japanese population, and therefore, its global applicability may be limited. Despite these limitations, EXPLORE-J is the largest registry study to investigate the diagnosis of FH using Achilles tendon imaging, and it provides valuable insights into the diagnosis of FH among ACS patients. This study showed that the prevalence of FH is at least five-fold higher in patients with ACS than in the general population. Prevalence is especially high in younger patients with premature onset of ACS. Timely screening and diagnosis, as well as aggressive lipidlowering therapy, are necessary for the successful management of FH and associated CAD. Results from EXPLORE-J provide critical insights that can enable an increased pace of diagnosis and treatment for patients with FH in Japan. Conflicts of interest MHS received honoraria from Astellas Amgen, Astellas, Sanofi, Aegerion, Kaneka Kowa, and MSD and research grants from Astellas Amgen, Astellas, Sanofi, Aegerion, and MSD. JA received honoraria from Sanofi and Amgen. AH received honorarium and research grant from Sanofi. AN received honorarium from Sanofi. AO and KU are employees of Sanofi. MN received honoraria from Sanofi and Astellas Amgen and research grant from Sanofi. HA and YM do not have any conflicts of interest to declare. Financial support This study was sponsored by Sanofi and Regeneron Pharmaceuticals, Inc. Author contributions Mariko Harada-Shiba, Junya Ako, Hidenori Arai, Atsushi Hirayama, Atsushi Nohara, and Masato Nakamura all served on the steering committee as principal investigators and contributed equally to conception and design of the study, protocol development, acquisition of data and interpretation of the data, and revising the publication for important intellectual content. Yoshitaka Murakami contributed to analysis and interpretation of the data and revising the publication for important intellectual content. Kiyoko Uno contributed to the design and implementation of the study, acquisition of data, interpretation of the data, and drafting and revising the publication for important intellectual content. Asuka Ozaki contributed to the acquisition of data, interpretation of the data, and drafting and revising the publication for important intellectual content. Mariko Harada-Shiba, Junya Ako, Hidenori Arai, Atsushi Hirayama, Yoshitaka Murakami, Atsushi Nohara, Asuka Ozaki, Kiyoko Uno, and Masato Nakamura approved the final version of the manuscript and have agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately
368
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
investigated and resolved. Acknowledgments The authors thank Tamio Teramoto of Teikyo University; Shun Ishibashi of Jichi Medical University; Kotaro Yokote of Chiba University; Tomonori Okamura of Keio University; and Hiroyuki Daida of Juntendo University for collaboration and advice. The authors would also like to thank Azusa Tsukida and Yuki Tajima of Sanofi for providing support with statistical analysis, Yosuke Ujike and Yuki Tajima of Sanofi for providing medical writing support, Yasuyoshi Nakahigashi of Sanofi and Mebix, Inc. for assistance with study implementation/operation, BML, Inc. for proprotein convertase subtilisin/kexin 9-related and genome-related analysis, Densuke Systems Co. Ltd. for statistical analysis, Shizuya Yamashita of Rinku General Medical Center, Toru Yoshizumi of Kawasaki Hospital, and Micron Inc. for radiography of Achilles tendons, and Regeneron Pharmaceuticals, Inc. for critical review of the manuscript. Writing and editorial assistance was provided by Ruhi Ubale, PhD, of Cactus Communications and was funded by Sanofi and Regeneron Pharmaceuticals, Inc.
[6]
[7]
[8]
[9]
[10] [11] [12]
[13]
References [14] [1] M. Harada-Shiba, H. Arai, S. Oikawa, T. Ohta, T. Okada, et al., Guidelines for the management of familial hypercholesterolemia, J. Atherosclerosis Thromb. 19 (2012) 1043e1060. [2] M. Harada-Shiba, H. Arai, T. Okamura, K. Yokote, S. Oikawa, et al., Multicenter study to determine the diagnosis criteria of heterozygous familial hypercholesterolemia in Japan, J. Atherosclerosis Thromb. 19 (2012) 1019e1026. [3] T. Teramoto, J. Sasaki, S. Ishibashi, S. Birou, H. Daida, et al., Familial hypercholesterolemia: executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan-2012 version, J. Atherosclerosis Thromb. 21 (2014) 6e10. [4] L.E. Akioyamen, J. Genest, S.D. Shan, R.L. Reel, J.M. Albaum, et al., Estimating the prevalence of heterozygous familial hypercholesterolaemia: a systematic review and meta-analysis, Br. Med. J. Open 7 (2017), e016461. [5] M. Benn, G.F. Watts, A. Tybjaerg-Hansen, B.G. Nordestgaard, Familial
[15]
[16] [17]
[18]
hypercholesterolemia in the Danish general population: prevalence, coronary artery disease, and cholesterol-lowering medication, J. Clin. Endocrinol. Metab. 97 (2012) 3956e3964. H. Mabuchi, A. Nohara, T. Noguchi, J. Kobayashi, M.A. Kawashiri, et al., Molecular genetic epidemiology of homozygous familial hypercholesterolemia in the Hokuriku district of Japan, Atherosclerosis 214 (2011) 404e407. B.G. Nordestgaard, M.J. Chapman, S.E. Humphries, H.N. Ginsberg, L. Masana, et al., Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society, Eur. Heart J. 34 (2013) 3478e3490. J. Pang, M. Hu, J. Lin, T. Miida, H.M. Nawawi, et al., An enquiry based on a standardised questionnaire into knowledge, awareness and preferences concerning the care of familial hypercholesterolaemia among primary care physicians in the Asia-Pacific region: the “Ten Countries Study”, Br. Med. J. Open 7 (2017), e017817. T. Teramoto, J. Sasaki, S. Ishibashi, S. Birou, H. Daida, et al., Executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan -2012 version, J. Atherosclerosis Thromb. 20 (2013) 517e523. B. Gencer, D. Nanchen, Identifying familial hypercholesterolemia in acute coronary syndrome, Curr. Opin. Lipidol. 27 (2016) 375e381. D. Mant, Commentary: what's so special about familial hypercholesterolaemia? Br. J. Gen. Pract. 59 (2009) 777e778. €ber, G.G. Stefanini, et al., Prevalence and D. Nanchen, B. Gencer, R. Auer, L. Ra management of familial hypercholesterolaemia in patients with acute coronary syndromes, Eur. Heart J. 36 (2015) 2438e2445. H. Ohmura, Y. Fukushima, A. Mizuno, K. Niwa, Y. Kobayashi, et al., Estimated prevalence of heterozygous familial hypercholesterolemia in patients with acute coronary syndrome, Int. Heart J. 58 (2017) 88e94. M. Nakamura, K. Uno, A. Hirayama, J. Ako, A. Nohara, et al., Exploration into lipid management and persistent risk in patients hospitalised for acute coronary syndrome in Japan (EXPLORE-J): protocol for a prospective observational study, Br. Med. J. Open 7 (2017), e014427. H. Mabuchi, S. Ito, T. Haba, K. Ueda, R. Ueda, Discrimination of familial hypercholesterolemia and secondary hypercholesterolemia by Achilles' tendon thickness, Atherosclerosis 28 (1977) 61e68. R.S. Rosenson, Myocardial injury: the acute phase response and lipoprotein metabolism, J. Am. Coll. Cardiol. 22 (1993) 933e940. A.C. Goldberg, P.N. Hopkins, P.P. Toth, C.M. Ballantyne, D.J. Rader, et al., Familial hypercholesterolemia: screening, diagnosis and management of pediatric and adult patients: clinical guidance from the National Lipid Association Expert Panel on Familial Hypercholesterolemia, J. Clin. Lipidol. 5 (2011) S1eS8. H. Yazici, Beware of registries for their biases, Bull. NYU Hosp. Joint Dis. 70 (2012) 95e98.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Prevalence of familial hypercholesterolemia in patients with acute coronary syndrome in Japan: Results of the EXPLORE-J study Mariko Harada-Shiba a, *, Junya Ako b, Hidenori Arai c, Atsushi Hirayama d, Yoshitaka Murakami e, Atsushi Nohara f, Asuka Ozaki g, Kiyoko Uno g, Masato Nakamura h a Department of Molecular Innovation in Lipidology, National Cerebral and Cardiovascular Center Research Institute, 5-7-1 Fujishiro-dai, Suita, Osaka, 5658565, Japan b Department of Cardiovascular Medicine, Kitasato University, 1-15-1 Kitasato, Minami-ku, Sagamihara, Kanagawa, 252-0375, Japan c Department of Geriatric Medicine, National Center for Geriatrics and Gerontology, 7-430 Morioka-cho, Obu City, Aichi, 474-8511, Japan d Division of Cardiology, Nihon University School of Medicine, 30-1 Oyaguchi Kamicho, Itabashi-ku, Tokyo, 173-8610, Japan e Department of Medical Statistics, Toho University, 5-21-16 Omori-Nishi, Ota-ku, Tokyo, 143-8540, Japan f Department of Cardiology, Kanazawa University of Graduate School of Medical Sciences, 13-1 Takara-machi, Kanazawa, Ishikawa, 920-8641, Japan g Sanofi, Tokyo Opera City Tower, 3-20-2 Nishi Shinjuku, Tokyo, 163-1488, Japan h Division of Cardiovascular Medicine, Toho University Ohashi Medical Center, 2-17-6 Ohashi Meguro, Tokyo, 153-8515, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 30 March 2018 Received in revised form 30 May 2018 Accepted 14 June 2018
Background and aims: Prevalence of familial hypercholesterolemia (FH), a common genetic disorder with a high risk for coronary artery disease (CAD), is high among CAD patients; however, data on FH prevalence among acute coronary syndrome (ACS) patients are limited. EXPLORE-J is the largest registry to diagnose FH among Japanese ACS patients using the 2012 Japan Atherosclerosis Society guidelines. Methods: This prospective study consecutively recruited patients between April 2015 and August 2016 at 59 sites. Low-density lipoprotein cholesterol (LDL-C) levels, family history of premature CAD, presence of tendon xanthomas, and Achilles tendon radiograms were recorded at baseline. The prevalence rate of FH in patients with ACS was estimated with 95% CI. Results: Of 1944 analyzed patients (mean age, 66.0 years; men, 80.3%), 52 (2.7% [95% CI: 2.0e3.5]) had FH. Thirty-one (1.6%) had LDL-C 180 mg/dL and Achilles tendon thickness (ATT) 9 mm, 8 (0.4%) had LDL-C 180 mg/dL and family history of premature CAD, 10 (0.5%) had ATT 9 mm and family history of premature CAD, and 3 (0.2%) met all the criteria. FH patients were younger than those without FH (59.5 [12.5] vs. 66.2 [12.1] years; p < 0.001). More patients with premature ACS (men, <55 years; women, <65 years) than without (4.7% [95% CI: 2.9e7.2] vs. 2.1% [1.4e3.0]) had FH. Conclusions: FH prevalence is at least five-fold higher in ACS patients than in the general population, especially in patients with premature ACS onset and ATT 9 mm. FH screening in ACS patients is therefore clinically important and critical. © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Familial hypercholesterolemia Achilles tendon thickness Acute coronary syndrome Coronary artery disease Low-density lipoprotein
1. Introduction Familial hypercholesterolemia (FH) is an inherited, autosomal
* Corresponding author. Department of Molecular Innovation in Lipidology, National Cerebral and Cardiovascular Center Research Institute, 5-7-1 Fujishiro-dai, Suita, Osaka, 565-8565, Japan. E-mail addresses: [email protected] (M. Harada-Shiba), [email protected] (J. Ako), [email protected] (H. Arai), [email protected] (A. Hirayama), [email protected] (Y. Murakami), [email protected] (A. Nohara), Asuka.Ozaki@sanofi.com (A. Ozaki), Kiyoko.Uno@sanofi.com (K. Uno), [email protected] (M. Nakamura).
dominant disease caused by abnormalities in the genes that code for low-density lipoprotein (LDL) receptors and related molecules. It is characterized by hyper-LDL cholesterolemia, early-onset coronary artery disease (CAD), and tendon or cutaneous xanthomas [1e3]. Globally, the estimated prevalence of FH is approximately 1 in 500 [3], although recent reports have estimated a prevalence of 1 in 250 [4], and even as high as 1 in 223 in a study in a Danish population [5]. The estimated number of cases in Japan is 300,000 [3], although a study conducted in the Hokuriku district of Japan reported a prevalence as high as 1 in 208 [6]. However, less than 1% of people affected by FH in Japan are estimated to receive a diagnosis;
https://doi.org/10.1016/j.atherosclerosis.2018.06.856 0021-9150/© 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
this problem of underdiagnosis may stem from the fact that patients with FH can be easily overlooked among the vast majority of individuals with cardiovascular disease caused by more common risk factors and by lack of knowledge among clinicians [7]. A survey conducted among primary care physicians worldwide showed that awareness of FH guidelines was low in Western countries such as the UK (61%), but was even lower in Japan (47%); this may possibly explain the underdiagnoses of FH in Japan [8]. The 2012 Japan Atherosclerosis Society (JAS) guidelines were more relaxed than most global guidelines and aimed for an LDL cholesterol (LDL-C) level of less than 100 mg/dL (2.6 mmol/L) or a 50% decrease in pretreatment LDL-C levels in patients with FH for primary prevention because of insufficient evidence to demonstrate that lower levels of LDL-C are required to reduce the risk of CAD in Japanese patients [9]. In the absence of stringent guidelines for LDL-C reduction in Japan and increased westernization of the Japanese lifestyle, the risk for CAD faced by patients with FH is high, and an efficient and accurate set of diagnostic criteria for FH are needed for early diagnosis and mitigation of risk for cardiovascular disease. According to the 2012 JAS guidelines for the diagnosis of FH, an adult (15 years) meeting two or more of the following criteria is diagnosed with FH: hyper-LDL cholesterolemia (untreated LDL-C level 180 mg/dL [4.7 mmol/L]); tendon xanthomas (including Achilles tendon hypertrophy [defined as Achilles tendon thickness (ATT) 9 mm]) or xanthoma tuberosum; and family history of FH or premature CAD (<55 years in men and <65 years in women) within first- and second-degree relatives. In addition, the diagnosis of FH should be made after eliminating the possibility of secondary hyperlipidemia [3]. It is important to note that, unlike any other set of criteria in any other country, JAS guidelines integrate the measurement of ATT, an easily accessible and interpretable modality, into the diagnosis of FH. In many patients, the initial clinical manifestation of FH is acute coronary syndrome (ACS) [10] that usually occurs at a younger age due to exposure to very high LDL-C levels from birth [11]. Therefore, the proportion of patients with FH is expected to be greater in patients with ACS than in the general population. A prospective, observational cohort study in patients hospitalized for ACS in Switzerland reported that according to the Dutch Lipid Clinic Network algorithm, 1.6% (95% confidence interval [CI]: 1.3e2.0%) had probable/definite FH and 17.8% (16.8e18.9%) had possible FH, whereas according to the Simon Broome Register algorithm, 5.4% (4.8e6.1%) had possible FH [12]. Ohmura et al. reported the results of a prospective, multicenter registry study to estimate the prevalence of FH in patients with ACS. Overall, 5.7% of patients with ACS were diagnosed with FH. Of those patients with ATT 9 mm, approximately 28.6% had FH [13]. However, because this study was limited by its small sample size, these results required confirmation in a larger population. EXPLORE-J is a prospective, large-scale, observational study of patients presenting with ACS conducted at 59 sites in Japan. To date, it is the largest registry study in Japan to focus on the high-risk ACS population and use ATT data for diagnosis of FH [14]. The objective of our current analysis was to describe the prevalence of FH in Japanese patients with ACS using the 2012 JAS guidelines and data from the EXPLORE-J registry study. 2. Patients and methods 2.1. Study design In this prospective, observational study conducted at 59 sites across Japan, patients who presented with ACS were consecutively enrolled between April 2015 and August 2016. Details of the study
363
design have been previously reported [14]. Briefly, patients were registered within 7 days following hospitalization for ACS. Data were collected at visits 1 through 5 during the 2-year observation period using an electronic case report form (Fig. 1). This study was conducted in compliance with the Declaration of Helsinki (amended in October 2013) and the Ethical Guidelines for Medical and Health Research Involving Human Subjects (enacted on 22 December 2014), and was approved by an ethical review committee at each site. All patients provided written informed consent prior to participation. 2.2. Patients Detailed inclusion and exclusion criteria have been previously reported [14]. Briefly, patients aged 20 years and over who present with ACS, including ST elevation myocardial infarction (STEMI), noneST elevation myocardial infarction (NSTEMI), or unstable angina, at the time of informed consent were eligible for inclusion. Patients with chest pain caused by prespecified comorbidities or instent thrombosis, patients enrolled in other interventional studies that could affect the lipid profile, and patients with conditions that could affect the monitoring and the natural course of the disease condition after treatment were excluded. 2.3. Assessments The highest available measurement of LDL-C obtained at or before visit 1 was used for analysis of FH. The LDL-C measurement (regardless of method) prior to hospitalization without treatment and the first measurement after hospitalization were reported. LDL-C levels were checked at visit 1 (within 14 days of hospitalization for ACS) and at every subsequent visit over the 2-year observation schedule (including both direct and calculated measurements). As previously described [15], ATT was measured at each site and at the central reading laboratory by three highly trained readers who were blinded to the patients' clinical characteristics using radiographs taken during hospitalization for registration in the study; however, radiographic measurement performed by visit 3 was also acceptable. Information on family history of CAD, ischemic cerebral infarction, and hypercholesterolemia within first- and second-degree relatives was collected at enrollment. Only family history of premature CAD was used for FH diagnosis. 2.4. Statistical analysis Details of the statistical analysis in this study have been previously reported [14]. Sample size was calculated to assess the persistent cardiovascular risk (major adverse cardiac events [MACE]) throughout the 2-year observation period, beginning with the index event. Based on the PACIFIC registry, in which the incidence of MACE was 6.4% at 2 years, a sample size of 2000 patients yielded a precision of ±1% in the incidence of MACE with a 95% CI of 0.053e0.074. Demographic variables were described by mean, median, standard deviation (SD), and range for continuous data, and by proportion in each category for categorical data. Fisher's exact test or Mann-Whitney U test was used for comparison of categorical variables according to nominal or ordinal scale. Student's t-test or Wilcoxon rank sum test was used for continuous variables as appropriate. A two-sided p value of <0.05 was considered significant. The prevalence of FH in patients with ACS was estimated with the associated 95% CI.
364
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
Fig. 1. Study design. a Patients were registered within 7 days of ACS. bRadiographs of the Achilles tendon will be obtained during hospitalization for registration as a rule, but a radiograph obtained by visit 3 is acceptable. ACS: acute coronary syndrome; LDL-C: low-density lipoprotein cholesterol; V: visit.
3. Results 3.1. Demographics and baseline characteristics Overall, 2016 patients were registered; of these, 1944 were included in the analysis (reasons for exclusion from the analysis population: failure to obtain informed consent within 7 days of hospitalization, 62; unapproved informed consent, 4; no informed consent obtained, 2; duplicate entry, 2; erroneous entry, 1; withdrawal due to achievement of target sample size, 1). The mean (SD) age was 66.0 (12.2) years, and the mean (SD) body mass index (BMI) was 24.2 (3.6) kg/m2; 80.3% of patients were men (Table 1). Cardiovascular risk factors such as hypertension (73.4%), dyslipidemia (77.8%), and diabetes (34.9%) were common. Overall, 625 (32.2%) patients were on lipid-modifying therapy (LMT): statins (530 [27.3%]), intensive statins (30 [5.7%]), eicosapentaenoic acid/docosahexaenoic acid (70 [3.6%]), ezetimibe (40 [2.1%]), and fibrates (35 [1.8%]).
for xanthomas, but family history of premature CAD was missing). On exclusion of patients with any missing data (n ¼ 1579), the diagnosis rate for FH was 3.1% (95% CI: 2.3e4.1). 3.3. Comparison of background data/baseline characteristics Baseline characteristics except age were similar between patients diagnosed with FH and those not diagnosed with FH. The mean (SD) age was 59.5 (12.5) years in patients diagnosed with FH versus 66.2 (12.1) years in patients without FH (p < 0.001) (Table 1). Proprotein convertase subtilisin/kexin 9 levels were comparable between patients diagnosed with FH and without FH. A lower proportion of patients with FH diagnosis than those without FH previously used LMTs (17.3% vs. 32.6%, p ¼ 0.023). However, among patients who had previously used statins, none of the seven patients diagnosed with FH and 30/523 (5.7%) without a diagnosis of FH were on intensive statins. 3.4. Subgroup analysis by particular diagnostic criterion for FH
3.2. Diagnosis of FH Of the 1944 analyzed patients, 52 (2.7% [95% CI: 2.0e3.5]) met two or more of the JAS criteria and were diagnosed with FH (Fig. 2). Most patients diagnosed with FH met only two criteria (49/1944 [2.5%]). Of these, 31/1944 (1.6%) had LDL-C 180 mg/dL and ATT 9 mm; 8/1944 (0.4%) had LDL-C 180 mg/dL and family history of premature CAD; and 10/1944 (0.5%) had ATT 9 mm and family history of premature CAD. Only 3/1944 (0.2%) patients met all three diagnostic criteria for FH. Of note, 19.3% of the patients met one of the three criteria. Of the analyzed patients, FH was indeterminable for 48 patients due to missing data (e.g., positive for LDL-C 180 mg/dL, negative
Overall, 23.2% (42/181) of patients who had LDL-C 180 mg/dL, 29.1% (44/151) of those who had xanthomas, and 13.9% (21/151) of those with family history of premature CAD were diagnosed with FH (Table 2). Data on family history of premature CAD were missing in 355 patients (18.3%), and this was the most common missing information. 3.5. Subgroup analysis by baseline characteristics A subgroup analysis of patients with FH according to baseline characteristics was conducted (Table 3). The proportion of patients with premature ACS onset (men, <55 years; women, <65 years)
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
365
Table 1 Demographics and baseline characteristics. All patients
Age (years), mean (SD) Men, n (%) BMI (kg/m2), mean (SD) PCSK9ad (ng/mL), median (Min, Max) LDL-C (mg/dL), mean (SD) Highest available level at baseline (1, 2, 3, or 4) 1) Measurement without medication prior to hospitalization 2) First measurement after hospitalization 3) Direct method 4) Calculated ACS type, n (%) STEMI NSTEMI Unstable angina Medical history, n (%) Coronary artery diseasew Cerebrovascular accidentw Peripheral artery diseasew Diabetes mellitus Hypertension Dyslipidemia Therapy before hospitalization Any LMT, n (%) Statin Intensive statiny,z Fibrates EPA/DHA Ezetimibe Antiglycemic (except insulin) Antiglycemic (insulin)
All patients (n ¼ 1944)
FH diagnosis (n ¼ 52)
No FH diagnosis (n ¼ 1892)
p value
66.0 (12.2) 1561 (80.3) 24.2 (3.6)c 356.0 (103.0, 1158.0)e
59.5 (12.5) 41 (78.8) 25.1 (3.6) 373.0 (165.0, 810.0)f
66.2 (12.1) 1520 (80.3) 24.2 (3.6)d 354.0 (103.0, 1158.0)g
<0.001a 0.727b 0.066a 0.078h
127.2 (39.5)i 132.6 (38.2)l 121.3 (40.0)o 101.8 (33.2)q 99.4 (31.9)t
201.2 187.1 191.9 153.8 144.5
125.1 (37.0)j 130.5 (36.8)n 119.2 (37.3)p 100.4 (31.3)s 98.2 (30.1)v
<0.001k <0.001a <0.001k <0.001k <0.001k
1195 (61.5) 309 (15.9) 440 (22.6)
31 (59.6) 8 (15.4) 13 (25.0)
1164 (61.5) 301 (15.9) 427 (22.6)
0.901b
355 (18.3) 149 (7.7) 37 (1.9) 679 (34.9) 1427 (73.4) 1512 (77.8)
5 (9.6) 3 (5.8) 1 (1.9) 21 (40.4) 33 (63.5) 50 (96.2)
350 (18.5) 146 (7.7) 36 (1.9) 658 (34.8) 1394 (73.7) 1462 (77.3)
0.060x 0.571x 0.801x 0.461b 0.112b <0.001b
625 (32.2) 530 (27.3) 30 (5.7)aa 35 (1.8) 70 (3.6) 40 (2.1) 382 (19.7) 77 (4.0)
9 7 0 1 1 2 7 1
616 (32.6) 523 (27.6) 30 (5.7)ac 34 (1.8) 69 (3.6) 38 (2.0) 375 (19.8) 76 (4.0)
0.023b 0.026b 1.000b 0.616b 1.000b 0.290b 0.293b 0.721b
(52.9) (36.1)m (58.9) (53.8)r (54.1)u
(17.3) (13.5) (0)ab (1.9) (1.9) (3.8) (13.5) (1.9)
ACS: acute coronary syndrome; BMI: body mass index; EPA/DHA: eicosapentaenoic acid/docosahexaenoic acid; FH: familial hypercholesterolemia; LDL-C: low-density lipoprotein cholesterol; LMT: lipid-modifying therapy; NSTEMI: noneST elevation myocardial infarction; PCSK9: proprotein convertase subtilisin/kexin 9; SD: standard deviation; STEMI: ST elevation myocardial infarction. a Student's t-test. b Fisher's exact test. c n ¼ 1937. d n ¼ 1885. e n ¼ 1876. f n ¼ 51. g n ¼ 1825. h Wilcoxon rank sum test. i n ¼ 1924. j n ¼ 1872. k Welch's t-test. l n ¼ 380. m n ¼ 14. n n ¼ 366. o n ¼ 1827. p n ¼ 1775. q n ¼ 1439. r n ¼ 39. s n ¼ 1400. t n ¼ 1797. u n ¼ 48. v n ¼ 1749. w Data were taken as Yes, No, or Possible; data for Yes are used in this table. x Mann-Whitney U test. y In patients previously treated with statins. z Intensive statins ¼ atorvastatin 20 mg, rosuvastatin 10 mg, and pitavastatin 4 mg. aa n ¼ 530. ab n ¼ 7. ac n ¼ 523. ad Heterodimer.
diagnosed with FH was higher than the proportion of patients without premature ACS onset and an FH diagnosis (4.7% [95% CI: 2.9e7.2] vs. 2.1% [95% CI: 1.4e3.0]). The prevalence of FH was 8.3% (95% CI: 1.8e22.5) in patients younger than 40 years and 2.6% (95% CI: 1.9e3.4) in those aged 40 years and older. FH was more
frequently diagnosed in patients without prior use of statins than in patients who had previously used statins (3.2% [95% CI: 2.3e4.2] vs. 1.3% [95% CI: 0.5e2.7]). There was no difference in FH prevalence between gender or ACS type.
366
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
Fig. 2. Diagnosis of FH using central review data of ATT. a Tendon xanthomas or ATT 9 mm. ATT: Achilles tendon thickness; CAD: coronary artery disease; FH: familial hypercholesterolemia; LDL-C: low-density lipoprotein cholesterol.
Table 2 Subgroup analysis of patients meeting particular diagnostic criterion for FH. Criteria
All patients All
LDL-C 180 mg/dL <180 mg/dL No data Tendon xanthomas Yes No Physical exam, Yes Physical exam, No ATT 9 mm (central reading) ATT <9 mm (central reading) Family history of premature CAD Yes No No data
Table 3 Subgroup analysis of prevalence rate by baseline characteristics.
FH diagnosis
All patients No FH diagnosis
181 1743 20
42 (23.2) 10 (0.6) 0 (0)
139 (76.8) 1733 (99.4) 20 (100.0)
151 1793 47 1897 116 1624
44 (29.1) 8 (0.4) 13 (27.7) 39 (2.1) 37 (31.9) 15 (0.9)
107 (70.9) 1785 (99.6) 34 (72.3) 1858 (97.9) 79 (68.1) 1609 (99.1)
151 1438 355
21 (13.9) 28 (1.9) 3 (0.8)
130 (86.1) 1410 (98.1) 352 (99.2)
All data are presented as n (%). ATT: Achilles tendon thickness; CAD: coronary artery disease; FH: familial hypercholesterolemia; LDL-C: low-density lipoprotein cholesterol.
4. Discussion To our knowledge, EXPLORE-J is the largest registry study conducted in Japan to evaluate the prevalence of FH using ATT measurement in patients with ACS. We observed that 2.7% of all patients with ACS included in the study met the JAS diagnostic criteria for FH. This rate of FH diagnosis among patients with ACS was at least five times higher than in the general population (0.2e0.5%) [3,6]. While the prevalence of FH observed in this analysis was lower than a previous study in Japan that reported a 5.7% prevalence of FH in patients with ACS [13], it was higher than a Swiss study that reported 1.6% using the Dutch Lipid Clinic Network algorithm [12]. The difference in the rate of diagnosis may have been affected by differences in the diagnostic criteria and nature of datasets. In this study, patients with FH were younger than those without FH (59.5 vs. 66.2 years; p < 0.001). Approximately 5% of patients with premature ACS onset were diagnosed with FH; this finding was comparable to a previous Swiss study wherein 4.8% of patients
All patients Sex Men Women ACS onset Premature Not premature <40 years 40 years ACS onset (men) Premature (<55 years) Not premature (55 years) ACS onset (women) Premature (<65 years) Not premature (65 years) ACS type STEMI NSTEMI Unstable angina Previous use of statins Yes No
All
FH diagnosis
Prevalence [95% CI]
1944
52
2.7 [2.0e3.5]
1561 383
41 11
2.6 [1.9e3.5] 2.9 [1.4e5.1]
426 1518 36 1908
20 32 3 49
4.7 2.1 8.3 2.6
340 1221
16 25
4.7 [2.7e7.5] 2.0 [1.3e3.0]
86 297
4 7
4.7 [1.3e11.5] 2.4 [1.0e4.8]
1195 309 440
31 8 13
2.6 [1.8e3.7] 2.6 [1.1e5.0] 3.0 [1.6e5.0]
530 1414
7 45
1.3 [0.5e2.7] 3.2 [2.3e4.2]
[2.9e7.2] [1.4e3.0] [1.8e22.5] [1.9e3.4]
All data are presented as n. ACS: acute coronary syndrome; CI: confidence interval; FH: familial hypercholesterolemia; NSTEMI: noneST elevation myocardial infarction; STEMI: ST elevation myocardial infarction.
with premature ACS onset also had FH [12], as well as a previous Japanese study in which 7.8% of patients with ACS under 60 years of age had FH [13]. In addition, the prevalence of FH diagnosis seems higher in patients under 40 years of age than in those 40 years and over (8.3% vs. 2.6%). This finding agrees with a previous study conducted in Japan [13], supporting the trend that FH is more common among patients with early onset of ACS, and this subpopulation should be one of the target patient groups for screening of FH. Although the JAS criteria have enabled more convenient diagnosis of FH, there are certain challenges associated with them. As previously mentioned, less than 1% of patients with FH in Japan are
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
estimated to receive an accurate diagnosis of FH [7]. A large proportion of patients with ACS use statins and other LMTs; in this study, 32.2% and 27.3% of patients had a history of use of any LMT and statin, respectively. While beneficial, LMTs may mask “true” LDL-C levels and tendon xanthomas in patients with FH and prevent accurate diagnosis. In addition, ATT is known to have a significant correlation with age [15]. Therefore, ATT in younger patients with FH may not be thick enough to meet the criteria, resulting in a missed diagnosis in such patients. It is important to note that young patients presenting with ACS are a warning sign for FH, and recording their family history of FH and premature CAD is extremely important for diagnosis. As mentioned previously, measurement of ATT for FH diagnosis is unique to this registry. Overall, 7.8% of the patients (151/1944) had tendon xanthomas, and among them, approximately one-third were diagnosed with FH (44 [29.1%]). This finding is similar to that of a previous study conducted in Japan, which reported that 28.3% of patients with ATT 9 mm also had FH [13]. There are two possible reasons for only one-third of patients with ATT 9 mm being diagnosed with FH. First, we did not have untreated LDL-C levels for all patients, and assuming that untreated LDL-C levels were higher than treated LDL-C levels, it is possible that patients with FH may have been misdiagnosed as not having FH because two of the three criteria for FH diagnosis were not met. Second, the criterion regarding FH diagnosis in patients with ATT 9 mm in the JAS guidelines is based on the study by Mabuchi et al. [15]; however, the patients in that study may not have used statins or other LMTs, unlike the patients in the current study. This may possibly have influenced the threshold of ATT 9 mm for FH diagnosis. Further studies may be needed to help us understand the appropriateness of using ATT as a diagnostic criterion for FH diagnosis in the general population or certain populations such as younger patients or patients with ACS. It has also been previously reported that LDL-C levels significantly decrease after an episode of acute myocardial infarction; LDL-C decreases by 31% on the day after the episode to as much as 48% on day 7 after the episode [16]. This means that diagnosis of FH using LDL-C levels just after an acute myocardial infarction could lead to underdiagnosis. Even though the benefits of LMT to lower LDL-C levels in FH patients are well known, reduced use of lipid-lowering treatment may be observed in Japan because until recently in the Japanese guidelines, a lower LDL-C target of less than 100 mg/dL (2.6 mmol/ L) was recommended [9]. In the current study, fewer patients diagnosed with FH reported previous use of statins than those without FH (13.5% vs. 27.6%). It may be inferred that patients who had not used statins previously had higher LDL-C and therefore met the diagnosis criteria for FH. Alternatively, it may be inferred that lower LDL-C levels in patients previously treated with statins may have possibly led to underestimation of FH prevalence. Nevertheless, patients with FH have been reported to have a 20-fold higher risk of CAD without treatment [17]; however, FH is treatable, and the risk of CAD can be reduced after a successful diagnosis. This requires aggressive lipid lowering to achieve LDL-C reduction of at least 50% or more [17]; therefore, an effort to ensure the timely diagnosis of FH followed by an aggressive treatment regimen is of the utmost importance for patients with FH in Japan. Some of the limitations associated with this study were the limited availability of data on untreated LDL-C levels and lack of data on family history of FH. Overall, untreated LDL-C data were only available in selected patients, including 380 patients with untreated LDL-C levels prior to hospitalization. As per the JAS guidelines, an untreated LDL-C level 180 mg/dL is a criterion for FH; therefore, assessment of untreated and treated LDL-C levels together could lead to an underestimation of FH prevalence. However, note that in patients for whom untreated LDL-C levels
367
prior to hospitalization were unavailable, the highest available measurement of LDL-C obtained at baseline, including the first measurement after hospitalization and the measurement at visit 1, was used for analysis of FH in this study. We believe that this may have helped to minimize underestimation of FH prevalence to some extent. Family history was particularly difficult to collect, and 355 patients had missing data after repeated efforts to collect data. With a numerically higher rate of FH prevalence when analysis was done for patients with data for all three diagnostic criteria available (3.1% vs. 2.7%), it is possible that the availability of all data would have provided us with a more accurate and potentially higher estimate of FH prevalence. Additionally, registry data are considered inherently susceptible to bias [18]. However, we avoided selection bias by enrolling consecutive patients. Finally, this study was restricted to the Japanese population, and therefore, its global applicability may be limited. Despite these limitations, EXPLORE-J is the largest registry study to investigate the diagnosis of FH using Achilles tendon imaging, and it provides valuable insights into the diagnosis of FH among ACS patients. This study showed that the prevalence of FH is at least five-fold higher in patients with ACS than in the general population. Prevalence is especially high in younger patients with premature onset of ACS. Timely screening and diagnosis, as well as aggressive lipidlowering therapy, are necessary for the successful management of FH and associated CAD. Results from EXPLORE-J provide critical insights that can enable an increased pace of diagnosis and treatment for patients with FH in Japan. Conflicts of interest MHS received honoraria from Astellas Amgen, Astellas, Sanofi, Aegerion, Kaneka Kowa, and MSD and research grants from Astellas Amgen, Astellas, Sanofi, Aegerion, and MSD. JA received honoraria from Sanofi and Amgen. AH received honorarium and research grant from Sanofi. AN received honorarium from Sanofi. AO and KU are employees of Sanofi. MN received honoraria from Sanofi and Astellas Amgen and research grant from Sanofi. HA and YM do not have any conflicts of interest to declare. Financial support This study was sponsored by Sanofi and Regeneron Pharmaceuticals, Inc. Author contributions Mariko Harada-Shiba, Junya Ako, Hidenori Arai, Atsushi Hirayama, Atsushi Nohara, and Masato Nakamura all served on the steering committee as principal investigators and contributed equally to conception and design of the study, protocol development, acquisition of data and interpretation of the data, and revising the publication for important intellectual content. Yoshitaka Murakami contributed to analysis and interpretation of the data and revising the publication for important intellectual content. Kiyoko Uno contributed to the design and implementation of the study, acquisition of data, interpretation of the data, and drafting and revising the publication for important intellectual content. Asuka Ozaki contributed to the acquisition of data, interpretation of the data, and drafting and revising the publication for important intellectual content. Mariko Harada-Shiba, Junya Ako, Hidenori Arai, Atsushi Hirayama, Yoshitaka Murakami, Atsushi Nohara, Asuka Ozaki, Kiyoko Uno, and Masato Nakamura approved the final version of the manuscript and have agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately
368
M. Harada-Shiba et al. / Atherosclerosis 277 (2018) 362e368
investigated and resolved. Acknowledgments The authors thank Tamio Teramoto of Teikyo University; Shun Ishibashi of Jichi Medical University; Kotaro Yokote of Chiba University; Tomonori Okamura of Keio University; and Hiroyuki Daida of Juntendo University for collaboration and advice. The authors would also like to thank Azusa Tsukida and Yuki Tajima of Sanofi for providing support with statistical analysis, Yosuke Ujike and Yuki Tajima of Sanofi for providing medical writing support, Yasuyoshi Nakahigashi of Sanofi and Mebix, Inc. for assistance with study implementation/operation, BML, Inc. for proprotein convertase subtilisin/kexin 9-related and genome-related analysis, Densuke Systems Co. Ltd. for statistical analysis, Shizuya Yamashita of Rinku General Medical Center, Toru Yoshizumi of Kawasaki Hospital, and Micron Inc. for radiography of Achilles tendons, and Regeneron Pharmaceuticals, Inc. for critical review of the manuscript. Writing and editorial assistance was provided by Ruhi Ubale, PhD, of Cactus Communications and was funded by Sanofi and Regeneron Pharmaceuticals, Inc.
[6]
[7]
[8]
[9]
[10] [11] [12]
[13]
References [14] [1] M. Harada-Shiba, H. Arai, S. Oikawa, T. Ohta, T. Okada, et al., Guidelines for the management of familial hypercholesterolemia, J. Atherosclerosis Thromb. 19 (2012) 1043e1060. [2] M. Harada-Shiba, H. Arai, T. Okamura, K. Yokote, S. Oikawa, et al., Multicenter study to determine the diagnosis criteria of heterozygous familial hypercholesterolemia in Japan, J. Atherosclerosis Thromb. 19 (2012) 1019e1026. [3] T. Teramoto, J. Sasaki, S. Ishibashi, S. Birou, H. Daida, et al., Familial hypercholesterolemia: executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan-2012 version, J. Atherosclerosis Thromb. 21 (2014) 6e10. [4] L.E. Akioyamen, J. Genest, S.D. Shan, R.L. Reel, J.M. Albaum, et al., Estimating the prevalence of heterozygous familial hypercholesterolaemia: a systematic review and meta-analysis, Br. Med. J. Open 7 (2017), e016461. [5] M. Benn, G.F. Watts, A. Tybjaerg-Hansen, B.G. Nordestgaard, Familial
[15]
[16] [17]
[18]
hypercholesterolemia in the Danish general population: prevalence, coronary artery disease, and cholesterol-lowering medication, J. Clin. Endocrinol. Metab. 97 (2012) 3956e3964. H. Mabuchi, A. Nohara, T. Noguchi, J. Kobayashi, M.A. Kawashiri, et al., Molecular genetic epidemiology of homozygous familial hypercholesterolemia in the Hokuriku district of Japan, Atherosclerosis 214 (2011) 404e407. B.G. Nordestgaard, M.J. Chapman, S.E. Humphries, H.N. Ginsberg, L. Masana, et al., Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society, Eur. Heart J. 34 (2013) 3478e3490. J. Pang, M. Hu, J. Lin, T. Miida, H.M. Nawawi, et al., An enquiry based on a standardised questionnaire into knowledge, awareness and preferences concerning the care of familial hypercholesterolaemia among primary care physicians in the Asia-Pacific region: the “Ten Countries Study”, Br. Med. J. Open 7 (2017), e017817. T. Teramoto, J. Sasaki, S. Ishibashi, S. Birou, H. Daida, et al., Executive summary of the Japan Atherosclerosis Society (JAS) guidelines for the diagnosis and prevention of atherosclerotic cardiovascular diseases in Japan -2012 version, J. Atherosclerosis Thromb. 20 (2013) 517e523. B. Gencer, D. Nanchen, Identifying familial hypercholesterolemia in acute coronary syndrome, Curr. Opin. Lipidol. 27 (2016) 375e381. D. Mant, Commentary: what's so special about familial hypercholesterolaemia? Br. J. Gen. Pract. 59 (2009) 777e778. €ber, G.G. Stefanini, et al., Prevalence and D. Nanchen, B. Gencer, R. Auer, L. Ra management of familial hypercholesterolaemia in patients with acute coronary syndromes, Eur. Heart J. 36 (2015) 2438e2445. H. Ohmura, Y. Fukushima, A. Mizuno, K. Niwa, Y. Kobayashi, et al., Estimated prevalence of heterozygous familial hypercholesterolemia in patients with acute coronary syndrome, Int. Heart J. 58 (2017) 88e94. M. Nakamura, K. Uno, A. Hirayama, J. Ako, A. Nohara, et al., Exploration into lipid management and persistent risk in patients hospitalised for acute coronary syndrome in Japan (EXPLORE-J): protocol for a prospective observational study, Br. Med. J. Open 7 (2017), e014427. H. Mabuchi, S. Ito, T. Haba, K. Ueda, R. Ueda, Discrimination of familial hypercholesterolemia and secondary hypercholesterolemia by Achilles' tendon thickness, Atherosclerosis 28 (1977) 61e68. R.S. Rosenson, Myocardial injury: the acute phase response and lipoprotein metabolism, J. Am. Coll. Cardiol. 22 (1993) 933e940. A.C. Goldberg, P.N. Hopkins, P.P. Toth, C.M. Ballantyne, D.J. Rader, et al., Familial hypercholesterolemia: screening, diagnosis and management of pediatric and adult patients: clinical guidance from the National Lipid Association Expert Panel on Familial Hypercholesterolemia, J. Clin. Lipidol. 5 (2011) S1eS8. H. Yazici, Beware of registries for their biases, Bull. NYU Hosp. Joint Dis. 70 (2012) 95e98.