Atherosclerosis Supplements 18 (2015) 154e162

www.elsevier.com/locate/atherosclerosis

Efficacy, safety, and tolerability of long-term lipoprotein apheresis in

patients with LDL- or Lp(a) hyperlipoproteinemia: Findings gathered
from more than 36,000 treatments at one center in Germany
Franz Heigl a,*, Reinhard Hettich a, Norbert Lotz a, Harduin Reeg a, Tobias Pflederer a,
Dirk Osterkorn b, Klaus Osterkorn b, Reinhard Klingel c
a
Dres. Heigl, Hettich, and Partner, Medical Care Center Kempten-Allgaeu, Robert-Weixler-Straße 19, 87439 Kempten, Germany
b
Medical Economics Institute, Zieblandstraße 9, 80799 Munich, Germany
c
Apheresis Research Institute, Stadtwaldgürtel 77, 50935 Cologne, Germany

Abstract

LDL cholesterol (LDL-C) and lipoprotein(a) (Lp(a)) are main risk factors for cardiovascular disease (CVD).
Efficacy, safety, and tolerability of lipoprotein apheresis (LA) were investigated in 36,745 LA treatments of 118 patients with CVD in
a retrospective, monocentric study. Indications were severe hypercholesterolemia (n ¼ 83) or isolated Lp(a) hyperlipoproteinemia (n ¼ 35).
Average age of patients at start of LA treatment was 58.1 years for males and 62.5 years for females. Medium interval between the first
cardiovascular event and LA treatment was 6.4  5.6 years and the average LA treatment period was 6.8  4.9 years. On average
treatments were performed once a week, via peripheral venous access in 79.3% of non-hemodialysis patients.
In patients with hypercholesterolemia initial pre-LA LDL-C was lowered from 176.4  67.0 mg/dL by 66.7  10.8% per session,
achieving a long-term interval mean value of 119.8  34.7 mg/dL, i.e. reduction by 32.1  19.6% (p < 0.0001). In patients with isolated
elevated Lp(a) initial pre-LA Lp(a) was lowered from 127.2  67.3 mg/dL by 66.8  5.8% per session, achieving a long-term interval
mean value of 60.0  19.5 mg/dL, i.e. reduction by 52.8  23.0% (p < 0.0001). After start of LA the average annual rate of major adverse
coronary events (MACE) of all patients declined by 79.7% (p < 0.0001). Subgroup analysis showed decline by 73.7% (p < 0.0001) in
patients with severe hypercholesterolemia, and by 90.4% (p < 0.0001) in patients with isolated elevated Lp(a). Adverse events (AE)
occurred in 1.1% of treatments.
LA treatment of patients with high risk for CVD due to LDL and/or Lp(a) hyperlipoproteinemia was effective, safe, and well tolerated.
The number of cardiovascular events, at least during a six-year period, declined by 80%.
Ó 2015 Elsevier Ireland Ltd. All rights reserved.

Keywords: LDL cholesterol; Lipoprotein(a); Cardiovascular events; Lipoprotein apheresis

1. Introduction
LDL-C has been recognized as most important risk
factor for coronary artery disease (CAD) for over thirty
* Corresponding author. Tel.: þ49 831 57057710; fax: þ49 831 570
577 51.
E-mail address: heigl@mvz-kempten.de (F. Heigl).
years [1]. Beginning with the 4S-study, it has been possible
in numerous clinical trials with statins and related meta-
analyses, to establish a clear link between lowering of
LDL-C and reduction of cardiovascular event rates [2,3]. In
recent years the equally atherogenic, thrombogenic, and
inflammatory potential of Lp(a), which was first identified
by K. Berg in 1963, has been established in epidemiolog-
ical studies [4e8]. According to Ariyo, an increase in the

http://dx.doi.org/10.1016/j.atherosclerosissup.2015.02.013
1567-5688/Ó 2015 Elsevier Ireland Ltd. All rights reserved.
 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162
Lp(a) level to 82 mg/dL is associated with an increase of
CAD by the factor 3 [9]. After withdrawal of nicotinic acid
in Europe in January 2013, there is no medication to
significantly lower an excessively high Lp(a) level. Unlike
with hypercholesterolemia, it was unclear for a long time
whether Lp(a) level reduction would positively influence
cardiovascular prognosis.
With LA treatment effective lowering of both LDL-C as
well as Lp(a) by 60%e80% is possible during a single
session. Systematic research in that field started with the
publication of Thompson et al. in 1975 on the successful
application of plasma exchange for the treatment of
homozygous familial hypercholesterolemia (FH) [10].
Further observation showed that the long-term application
of plasma exchange had a positive influence on the risk of
atherosclerosis of coronary vessels and other vascular
regions [11].
In the eighties and early nineties of the twentieth century
in order to replace unselective plasma exchange, several
selective LA methods were developed using physico-
chemical principals of precipitation, filtration, and adsorp-
tion to remove only those substances from plasma or blood
that cause atherosclerosis - in particular LDL-C and Lp(a)
[12e16]. For all LA methods it has been shown for
hypercholesterolemic patients that there was a long-term
effect of improved lipid metabolism. In conjunction with
hypothesized additional pleiotropic effects of LA rate of
major adverse coronary events (MACE) of chronic LA
patients is reduced by an average of 80% [17e19].
After encouraging observations of individual patients
with isolated Lp(a) hyperlipoproteinemia, the results of LA
treatment were published for this new indication in
a multicenter, longitudinal cohort study with 120 patients
[20]. Through an average reduction of the Lp(a) level by
36.3% from 117.9 mg/dL to 75.1 mg/dL with LA treatment,
the per year and per patient MACE count could be signif-
icantly reduced from 1.06 to 0.14, i.e. a reduction of 86%.
Due to methodological weaknesses in this study and
considering cost of LA reimbursement a prospective study
was stipulated. A randomized design of the study, which
was initially suggested, was rejected by ethics committees
in view of the favorable results of the retrospective study.
The multicenter study “Pro(a)LiFe” was published in
September 2013. Here 170 patients were included matching
the criteria for isolated elevated Lp(a) according to the
German reimbursement authority Federal Joint Committee
[21]. Over the course of two years of patient observation
before and after commencing LA treatment, a medium
reduction of the Lp(a) level was achieved from
104.9  45.7 mg/dL to 70.9  26.8 mg/dL, i.e. by 32.4%.
The annual per patient MACE rate declined from
0.41  0.45 to 0.09  0.22 meaning a significant reduction
of 78%. Overall, this prospective study was able to firmly
underpin the results of the earlier retrospective study.
In Germany, LA is covered by regular reimbursement of
statutory health insurance since 1991. Recognized
155
indications include: “patients with homozygous familial
hypercholesterolemia or patients with severe hypercholes-
terolemia, whose LDL-C could not be sufficiently reduced
over a documented twelve month course of medication and
prescribed diet.” The comprehensive risk profile has to be
considered for the indication of LA [22]. Since 2008, LA
additionally has been recognized as a treatment for
“patients with an isolated elevated Lp(a) level above 60 mg/
dL and LDL-C within the norm, and concomitant
progressive CVD (coronary, peripheral arterial, or cere-
brovascular) as assessed and documented clinically and by
imaging techniques.”
Currently, more than 2000 patients are being treated in
Germany within LA programs of about 300 medical
centers. Indications are homozygous FH app. 6% of
patients, severe hypercholesterolemia app. 64%, and iso-
lated elevated Lp(a) app. 30% [23]. LA treatments should
preferably be carried out in medical centers that practice
apheresis treatments on a large scale (>500 per year) while
maintaining an appropriate level of quality and entering
treatment data into a national, or ideally even an interna-
tional registry [24e28].
2. Patients and methodology
2.1. Study design, treatment location, and patient
recruitment
A monocentric, retrospective, longitudinal cohort study
was carried out at our medical competence center for
apheresis, performing nearly 6000 LA treatments per year.
All investigated patients were approved for chronic LA
treatment according to the Guidelines of the Joint Federal
Committee, following an initial and annually renewable
application, due to the following diagnoses: severe hyper-
cholesterolemia or isolated elevated Lp(a) with progressive
CVD.
The observation period covered October 1996 to
December 2013. Retrospective analysis of side effects was
based on the complete archives of treatment records of all
lipoprotein apheresis sessions of the center.
Following the examination of the study protocol by the
Bavarian State Medical Association, a letter was received
on January 21, 2014, which declared that the study could
proceed and be published without the approval of an ethics
committee, given that it was intended as a quality assurance
measure for the clinical treatment of patients in our center.
2.2. Lipoprotein apheresis
Selective LA methods used within this study period
were described in the literature and were performed
according to the instructions for use supplied by manu-
facturers [26]. Methods were: heparin-induced LDL
precipitation apheresis (H.E.L.P., Plasmat Futura: B. Braun,
Melsungen, Germany), polyacrylate adsorption from whole
156 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162
blood and simple double-filtration plasmapheresis (DALI
and MONET, Fresenius Medical Care, Bad Homburg,
Germany) and temperature-optimized double-filtration
plasmapheresis (DFPP, Lipidfiltration, Asai Kasei Medical,
Japan, and Octo Nova, Diamed Medizintechnik, Cologne,
Germany).
2.3. Laboratory measurements
Total cholesterol, LDL-C, HDL-C, triglycerides, Lp(a),
fibrinogen, hemoglobin and creatinine levels were
measured immediately preceding LA treatment quarterly in
all LA patients, as well as HbA1c levels in all diabetics.
Lipoprotein levels are expressed as mg/dL. LDL-C can be
converted to SI units following recommendations of the
American Medical Association Manual of Style, 10th ed.
from mg/dL to mmol/L by factor 0.0259. Methods for
measurement of Lp(a) changed over the entire 18 years
observation period but were all standardized for mass with
30 mg/dL as the upper limit of normal to adopt the
requirements of the German reimbursement guideline and
provided results without major variance regarding indi-
vidual patient courses. Conversion of Lp(a) levels into
a molar SI unit is not possible because the kringle 4 type 2
copy number of the patient would have to be known. Lipid
parameters, fibrinogen, and hemoglobin were then
measured again right after the treatment session.
Percentage reduction rates of LDL-C and Lp(a) were
calculated from levels directly preceding and following LA
treatment after complete return of patients’ blood. Regular
correction for putative hemodilution after LA, e.g. using
hematocrit, was not performed. LDL and Lp(a) levels rise
again between two apheresis treatments following a curve
linear kinetic. The interval mean value Cmean was calcu-
lated according to the formula suggested by Kroon [29]:
Cinterval mean ¼ Cmin þ 0,73 , (Cmax e Cmin), where Cmin
corresponds to the minimal LDL-C or Lp(a) concentrations
right after LA treatment and Cmax expresses the LDL-C or
Lp(a) concentrations immediately before the next LA
treatment. This empirical formula has been generated as
estimation of time-averaged levels of LDL-C and Lp(a)
from patients treated by biweekly regular LA. According to
the rebound curve patterns it represents the best available
approximation for weekly intervals.
2.4. Outcome parameters and statistics
The primary outcome parameter was the annual inci-
dence rate of cardiovascular events during chronic LA
treatment. In accord with other authors’ definitions [20,21],
cardiac death, a non-lethal myocardial infarction (MI),
coronary bypass surgery (CABG) as well as percutaneous
coronary intervention or stent implantation (PCI) all
counted as major adverse coronary events (MACE) repre-
senting the primary composite outcome parameter. Adverse
cardiac and vascular events (ACVE) made up the secondary
composite outcome parameter, meaning all coronary as
well as vascular events in all non-cardiac vascular regions.
Events or interventions relating to just a single described
location in the vascular system within 28 days were
considered a singular event. Any events relating to 2
described locations in the vascular system were considered
2 events with exception of multiple vessel CABG, which
was counted as just 1 intervention. The risk reduction rate
through LA treatment was calculated as the difference
between the rate of these events before (average observa-
tion period 6.4  5.6 years; range 1e27 years) and after
commencing chronic LA treatment (average observation
period 6.8  4.9 years; range 1e23 years).
Analysis of baseline characteristics and subsequent
status records was performed by the use of routine methods
of descriptive statistics. Two-sided paired Wilcoxon test
was used for MACE and ACVE rates. For each individual
patient, the absolute total number of events was used as
a variable for the two-sided paired Wilcoxon test. Differ-
ences of Lp(a) and LDL-C before and after LA treatments
were assessed by the paired Wilcoxon test.
3. Results
3.1. Patient characteristics
The study included a total of 118 consecutive patients,
who received chronic LA treatment between October 1996
and December 2013 at our center for a mean individual
period of 6.8  4.9 (range 1e23) years. This amounted to
a total of 797 treatment years. The total number of treat-
ment sessions amounted to 36,745. Patient characteristics
are summarized in Table 1. About two thirds of patients
were male and one third was female. Average age of males
at the beginning of LA treatment was 58.1 (range 37e81)
and for females it was 62.5 (25e83) years. The first cardiac
event of the patients had taken place 6.4  5.6 (range
1e27) years prior. 100% of the patients had CAD, 79.7%
also had cerebrovascular disease, and 22.9% peripheral
artery occlusive disease. The cardiovascular risk factors
were as follows: 67% had a positive family history, 61%
required medical treatment for arterial hypertension and
23.7% were diabetic. Only 2 patients were regular smokers
at the beginning of LA treatment. 7 patients (5.9%) had end
stage renal disease with hemodialysis. Indication for LA
treatment was severe hypercholesterolemia in app. 70%.
88% of these had intolerance to drug therapy, e.g. statin
myopathy, 12% exhibited ineffectiveness of medication.
App. 30% had an LA treatment due to the indication of
isolated elevated Lp(a) of >60 mg/dL with progressive
CVD. To fulfill the reimbursement criteria of isolated Lp(a)
hyperlipoproteinemia LDL-C levels were treated by maxi-
mally tolerated lipid lowering medication to reach the
target level of <100 mg/dL as part of optimized control of
all other cardiovascular risk factors. Independent of the
criteria for LA indication there was a substantial overlap
 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162 157

Table 1 Table 2
Baseline characteristics at time of first LA treatment. LA methods and treated plasma and blood volumes.
All patients (n ¼ 118) LA method Number of Plasma (P) and blood (B)
Male/female 77 (65.3%)/41 (34.7%) treatments volumes, mL per treatment
Age, years 59.6  11.2 (range 25e83) (n ¼ 36,745)
Male, years 58.1  10.5 (range 37e81) H.E.L.P.-apheresis 17,758 (48.3%) 3103  705 (P)
Female, years 62.5  11.9 (range 25e83) Temperature 9370 (25.5%) 3100  675 (P)
Body mass index, kg/m2 27.7  4.2 optimized DFPP
Cardiovascular risk factors: Polyacrylate 9218 (25.1%) 8129  1173 (B)
Positive family history 79 (67.0%) adsorption
Hypercholesterolemia 111 (94.1%) (DALI)
Lp(a) hyperlipoproteinemia 83 (70.3%) Simple DFPP 399 (1.1%) 3600  842 (P)
Arterial hypertension 72 (61.0%) (MONET)
Diabetes mellitus 28 (23.7%)
Smoking habits
Never 69 (58.5%)
Laboratory parameters before and during chronic LA
Former 47 (39.8%)
Current 2 (1.7%) treatments are summarized in Tables 3 and 4. Patients
Chronic renal failure as assessed by Cockcroft-Gault formula: eGFR, mL/ whose indication was severe hypercholesterolemia had an
min initial LDL-C value prior to chronic LA treatment of
30e59 18 (15.3%) 176.4  67.0 mg/dL and achieved reduction rates after LA
15e29 2 (1.7%)
with an average of 66.7%  10.8% (p < 0.0001). In long-
<15 or dialysis 7 (5.9%)
Cardiovascular diseases term LA treatment the interval mean value was
Coronary artery disease (CAD) 118 (100%) 119.8  34.7 mg/dL which was 32.1% below the initial
1-vessel-CAD 16 (13.5%) value (p < 0.0001). In this group 48 out of 83 patients
2-vessel-CAD 18 (15.3%) (57.8%) had also Lp(a) level > 30 mg/dL. This was on
3-vessel-CAD 84 (71.2%)
average 75.0 mg/dL which is about 2.5 times the upper
Cerebrovascular disease 94 (79.7%)
Peripheral artery disease 27 (22.9%) limit of normal. The upper average limit could be lowered
Indication for LA simultaneously with that of LDL-C with equal efficacy per
Severe hypercholesterolemia 83 (70.3%) session but by 56.4% regarding average levels of long-term
treatment failure 10 (12.0%) treatment. So the long-term Lp(a) reduction was substan-
intolerability of medical treatment 73 (88.0%)
tially higher compared to LDL-C reduction.
Isolated Lp(a) elevation 35 (29.7%)
Time between 1st cardiovascular event and 6.4  5.6 (range 1e27) In the patient group with isolated elevated Lp(a) the
1st LA, years level was more than four times the upper limit of normal
with 127.2 mg/dL, LDL-C was 96.1 mg/dL. Lp(a) and
LDL-C levels were lowered by 66.8% and 65.7% respec-
tively in a single LA session. During long-term LA treat-
between the two dyslipidemia types in the 118 patients: 111 ment the interval mean value of Lp(a) level was 60 mg/dL,
had general hypercholesterolemia (94.1%) and 83 patients which was a reduction by 52.8%, whereas average LDL-C
had Lp(a) hyperlipoproteinemia (70.3%). was just 22.7% lower than the initial value.
79.3% of non-hemodialysis patients were treated via
peripheral venous access whereas 20.7% required an AV
shunt. The average interval between treatments was 1 week 3.3. Analysis of events
resulting in a mean of 51.3 treatments per patient per year.
The distribution of apheresis methods and treatment Analysis of events was performed for 118 patients who
volumes of plasma and blood are summarized in Table 2. were observed for an average of 6.4  5.6 (range 1e27)
years between their first cardiovascular event and the
beginning of LA treatment and were then observed for an
3.2. Medication and laboratory parameters average of 6.8  4.9 (range 1e23) years during chronic LA
treatment. The absolute number of cardiovascular events
All patients were already receiving maximally tolerated (MACE and ACVE) preceding and following initiation of
lipid lowering medication and cardiac medication prior to chronic LA treatment and derived reduction rates for all
and during each stage of chronic LA treatment. Notably, investigated patients and for the patients of both treatment
these included platelet aggregation inhibitors, beta indications are shown in Table 5. During this period, the
blockers, ACE inhibitors and angiotensin receptor blockers average annual MACE rate per patient could be reduced
as well as diuretics. The contraindication of ACE inhibitors from 0.35 to 0.07 in the total group of all 118 patients. The
with adsorption techniques had to be considered. Besides average reduction rate after initiation of LA treatment was
hyperlipidemia all other cardiovascular risk factors were 79.7% for MACE and 73.3% for ACVE. In the group of
individually optimized. patients with isolated elevated Lp(a) reduction of the
158 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162

Table 3
Plasma concentrations of LDL-C and Lp(a) in patients with severe hypercholesterolemia before, in first month, and during steady state of chronic LA. (There
have been patients in whom no Lp(a) was detectable, but they haven’t been separately recorded.).
LDL-C Before LA 1st month Steady state Mean total LDL-C reduction [%]
Cmax before LA [mg/dL] not done 154.7  51.4 148.8  43.9
Cmin after LA [mg/dL] not done 63.1  26.8 48.9  18.6
Reduction by LA [%] not done 58.7  12.0 66.7  10.8 (p < 0.0001)
Interval mean level [mg/dL] 176.4  67.0 128.1  42.6 119.8  34.7 32.1  19.6 (p < 0.0001)
Lp(a) Before LA 1st month steady state Mean total Lp(a) reduction [%]
Cmax before LA [mg/dL] not done 52.5  55.4 40.2  38.9
Cmin after LA [mg/dL] not done 23.4  25.2 14.0  12.6
Reduction by LA [%] not done 54.6  20.7 65.2  30.7 (p < 0.0001)
Interval mean level [mg/dL] 75.0  64.5 48.3  47.2 32.7  31.0 56.4  20.5 (p < 0.0001)

Table 4
Plasma concentrations of Lp(a) and LDL-C in patients with isolated Lp(a) elevation before, in first month, and during steady state of chronic LA.
Lp(a) Before LA 1st month Steady state Mean total Lp(a) reduction [%]
Cmax before LA [mg/dL] not done 97.2  56.0 74.5  24.3
Cmin after LA [mg/dL] not done 39.6  23.1 24.5  8.9
Reduction by LA [%] not done 59.5  11.9 66.8  5.8 (p < 0.0001)
Interval mean level [mg/dL] 127.2  67.3 81.2  45.8 60.0  19.5 52.8  23.0 (p < 0.0001)
LDL-C Before LA 1st month steady state Mean total LDL-C reduction [%]
Cmax before LA [mg/dL] not done 92.6  33.9 91.7  27.9
Cmin after LA [mg/dL] not done 41.0  18.2 31.8  15.6
Reduction by LA [%] not done 55.3  12.9 65.7  8.7 (p < 0.0001)
Interval mean level [mg/dL] 96.1  33.5 77.7  28.1 74.3  23.7 22.7  31.2 (p < 0.0001)

MACE rate was 90.4% and 89.5% for ACVE, which patient per year. The MACE rate increased continuously
appeared higher than in the group of patients with severe prior to LA treatment from 0.04 in year 6 to 0.9 in year
hypercholesterolemia with 73.7% for MACE and 64.1% for 1. Once LA treatment began, this declined from 0.15 in
ACVE. year þ1 to 0.02 in year þ6. The number of analyzed
Fig. 1 shows the number of annual MACE rates in patients varied in years during LA treatment due to the
relation to the number of patients observed preceding and retrospective design. However, comparing corresponding
following LA treatment, that is to say the MACE rate per years before and during LA treatment within a 12 years
period (year 6 to year þ6) all pairs exhibited a MACE
reduction being essentially identical to the mean reduction
Table 5 calculated with all events for the entire observation period.
Total number of MACE and ACVE in 6.4  5.6 (range 1e27) years before
and 6.8  4.9 (range 1e23) years after commencing chronic LA.
3.4. Safety and tolerability of LA treatment
Before LA [n] During LA [n] Reduction rate [%]
All patients (n ¼ 118) Table 6 shows that during 36,745 LA treatments, there
MACE 261 53 79.7 (p < 0.0001)
MI 68 10 85.3 (p < 0.0001)
were unexpected adverse events in 1.1%, vascular problems
PCI 138 37 73.2 (p < 0.0001) in 2.1% and technical problems in 0.08%. Overall, 99.5%
CAGB 55 6 89.1 (p < 0.0001) of all treatments reached their treatment target, defined by
ACVE 289 77 73.3 (p < 0.0001) the plasma or blood volume intended to treat.
Patients with severe hypercholesterolemia (n ¼ 83)
MACE 167 44 73.7 (p < 0.0001)
MI 48 8 83.3 (p < 0.0001) 3.5. Termination of chronic LA treatment
PCI 81 31 61.7 (p < 0.0001)
CAGB 38 5 86.8 (p < 0.0001) LA treatment was terminated during the observation
ACVE 184 66 64.1 (p < 0.0001)
period for 27 of the total 118 patients. Reasons for termi-
Patients with isolated Lp(a) elevation (n ¼ 35)
MACE 94 9 90.4 (p < 0.0001) nation are summarized in Table 7. Chronic LA treatment
MI 20 2 90.0 (p < 0.0001) was halted in 4 patients at their request resulting in treat-
PCI 57 6 89.5 (p < 0.0001) ment adherence rate of 99.5% per year. Therapy was ceased
CAGB 17 1 94.1 (p < 0.0001) in 6 patients due to non-cardiac progressive malignant
ACVE 105 11 89.5 (p < 0.0001)
diseases. 15 patients died, amounting to 1 every 53
 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162 159

Fig. 1. MACE rates per year in relation to the number of observed patients and in relation to the chosen time interval. All patients [n ¼ 118] with severe
hypercholesterolemia [n ¼ 83] and isolated Lp(a) elevation [n ¼ 35] are included in this figure.

treatment years. 40% of those were dialysis patients. Less

Table 6 than half of the patients died from a cardiac event. This
Adverse events (AE), vascular access problems (VAP) and technical amounts to a lethal cardiac event rate of 8.8% every 10
problems (TP) in LA treatment (n ¼ 36,745). AE severity was graded in years in a high-risk patient population with a non-lethal 10
class I (mild AE not requiring any intervention), class II (therapeutic year event rate of 217% prior to the administering of LA
intervention or temporary break of LA session), and class III (termination
of LA session).
treatment. 256 cardiovascular events occurred in 118
patients within 10 years prior to commencing LA.
 Total number of AE: 413 (1.1%)
I 157 (38.0%)
II 159 (38.5%) 4. Discussion
III 97 (23.5%)
 Most frequent AE symptoms:
In this retrospective study 118 patients with established
e Hypotension 219 (53%)
e Prolonged bleeding or 126 (30%) indication for lipoprotein apheresis due to hyper-
hematoma after LA lipoproteinemia with high CVD risk were analyzed, who
e Bradycardia 25 (6%) received more than 36,000 LA treatment sessions for an
e Allergic skin or mucosa 15 (3.6%) average time period of nearly seven years. LA
reaction
e Edema 15 (3.6%) Table 7
e Nausea/vomiting 14 (3.4%) Reason for terminating chronic LA in the cohort of 118 patients covering
e Dizziness 6 (1.5%) 797 patient years.
 Change of LA method:
e Intolerability (allergic 9/118 patients (7.6%) Reason Number of patients Relation to total number
reactions) terminating LA of patient years
 Total number of VAP 774 (2.1%) Total 27
I 696 (89.9%) Death 15 1 per 53 patient years
II 5 (0.6%) Dialysis patients 6 1 per 4 patient years
III 73 (9.5%) Non-dialysis patients 9 1 per 86 patient years
 Total number of TP 29 (0.08%) Cardiac death 7 1 per 114 patient years
I 4 (13.8%) (8.8% per 10 patient
II 7 (24.1%) years)
III 18 (62.1%) Progressive malignant 6
 Treatments without any 35,529 (96.7%) disease
problem (AE, VAP or TP) Change of patient‘s 2
 Treatments successfully 36,557 (99.5%) residence
completed Patient‘s will 4 1 per 200 years
160 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162
demonstrated to be an effective, safe, and well tolerated
extracorporeal treatment. Both LDL-C as well as Lp(a)
levels could be lowered acutely by apheresis sessions by an
average of 66%, irrespective of the initial level. With saw-
tooth like changes of lipoproteins during regular LA
treatment a stronger rebound was observed for LDL-C than
for Lp(a) [21,29]. This difference in lipoprotein recovery
might be the explanation for the phenomenon that LA,
achieving identical reduction rates immediately after the
session for both LDL-C and Lp(a), in the long-term showed
significantly weaker LDL-C reduction of 32% in patients
with severe hypercholesterolemia compared to 53% long-
term Lp(a) reduction in patients with isolated elevated
Lp(a). Those patients who started Lp(a) apheresis treatment
with an already very low LDL-C level of 96 mg/dL, had an
even faster LDL-C rebound and a correspondingly smaller
long-term reduction rate of only 23%. A similar observa-
tion of 18% long-term LDL-C reduction was noted in the
Pro(a)LiFe study [21]. In accord with Jaeger et al. [20],
who reported a 36% long-term Lp(a) reduction, but in
contrast to our own findings (53%), authors of this study
also noted a lower long-term Lp(a) reduction of 32% with
initial values that were a little lower than in our patient
population: 127.4 mg/dL compared to 105 mg/dL [21] or
118 mg/dL [20].
In all three investigated populations lowering of MACE
rates following extracorporeal Lp(a) elimination was equally
impressive by 78% [21], by 86% [20], and by 90% respec-
tively. The initial risk level in the year prior to starting LA
treatment was very high, with an event rate of 0.54 in 170
patients [21], 1.06 in 120 patients [20] and 1.49 for the 35
Lp(a) patients in our study. The extent of cardiovascular risk
in these patient groups becomes clear in comparison to the
Framingham Risk Score [30] categorizing a high-risk patient
if the cardiac event rate over a 10 year period is  20%,
corresponding to an annual rate of 0.02. The apheresis
patients being discussed here showed, at least during the last
year before LA treatment, a risk estimate that was more than
50 times higher. The comparison between the three patient
populations allows to speculate whether it is reasonable to
expect that a patient who is at a higher cardiovascular risk
preceding LA treatment shows a stronger reduction in the
MACE rate during LA treatment. At this point, it should also
be noted that this purely retrospective study cannot be
compared directly with the prospective and clearly defined
Pro(a)LiFe study [21] since the observation periods
preceding and following LA treatment varied and patient
number was lower. The comparison between both LA indi-
cation groups, i.e. severe hypercholesterolemia or isolated
elevated Lp(a) is interesting as it both adds and confirms
clinical results in this field [21,31].
About 70% of LA patients across Germany are treated
due to severe hypercholesterolemia and 30% due to iso-
lated elevated Lp(a) [23], which fully matches our patient
population in proportion. The initial risk for our patients
with high LDL-C levels was markedly lower with 0.65 in
the year prior to LA treatment than for Lp(a) patients with
1.49. MACE reduction with LA treatment was lower in
LDL-C patients at 73.7% than in the Lp(a) group at
90.4%. This appears to confirm the conclusion that the
higher the initial risk, the greater the effect LA treatment
has. This relationship has already been observed between
statin therapy and hypercholesterolemia leading to the
number needed to treat. In the Pro(a)-LiFe patient pop-
ulation, only 3 patients needed LA treatment for 1 year in
order to avoid a MACE [21], which matches our findings.
Even if the risk reduction through LA was somewhat
smaller for hypercholesterolemia patients than for Lp(a)
patients, it is apparent that the 73.7% reduction of
cardiovascular events in relation to an average LDL-C
lowering of “just” 56 mg/dL meaning 32% is more than
double as high as expected from the LDL-C elimination by
statins alone [2,32]. Though we do not have much
comparable data, mechanisms should be further elucidated
how the huge risk reduction that is also observed in Lp(a)
patients can be reached inspite of an only relatively small
long-term Lp(a) reduction as described in recent studies
[20,21].
One explanation may be the simultaneous lowering of
Lp(a) and LDL-C representing 2 independent risk factors.
A counter argument for this would be the fact that for
precisely those patients with isolated elevated Lp(a) the
LDL-C level was already at a low level of around 100 mg/
dL before starting LA treatment, and then declined only
slightly once LA treatment has been started. It is also
difficult to explain that our patient group with isolated
elevated Lp(a) of 127 mg/dL and an a priori well controlled
LDL-C level of 96 mg/dL had a higher cardiovascular risk
and had greater benefit from LA than our patient group with
severe hypercholesterolemia and LDL-C of 176 mg/dL,
which simultaneously showed a relatively high average
Lp(a) level of 75 mg/dL in the overlap area between both
dyslipidemia types. This observation could also lead to
deduce that the Lp(a) level has a disproportionate influence
on ACVE, independent of the LDL-C level. The example of
the comparison of the two patient groups’ laboratory
parameters showed that the difference in the Lp(a) level of
52 mg/dL (127 mg/dL in the group with isolated Lp(a)
elevation and 75 mg/dL in the group with severe hyper-
cholesterolemia) has a greater prognostic significance than
the difference in the LDL-C level of 80 mg/dL (176 mg/dL
in the group with severe hypercholesterolemia and 96 mg/
dL in the group with Lp(a) elevation). Possibly the high risk
potential of Lp(a) can be explained by atherogenic and
additionally thrombogenic effects.
Additional to the lipoprotein eliminating effect of LA
a number of pleiotropic, especially pro-rheological and
anti-inflammatory effects must be mentioned [33e36].
Given the average LDL and Lp(a) level reductions, which
seem to not fully explain clinical efficacy it could be
hypothesized that the pulsatile fall in lipoproteins to
extremely low values immediately after LA treatment
 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162
might have a sustained positive effect on the vascular
endothelium.
The clinical efficacy of LA can also be measured
regarding patient mortality. Even if there are no mortality
data available that can be compared with the patient pop-
ulation being investigated here, it becomes apparent that the
cardiac mortality rate of just one death in 114 treatment
years on a patient group with an initial average age of 60
years and a very high risk for CVD is unexpectedly low.
Regarding the above-mentioned methodological weak-
ness of variable observation periods of the retrospective
analysis and the potential resulting errors in estimating
event rates, it should be noted that our analyses showed no
statistically significant difference if corresponding years
were compared during the observational period of 6 years
preceding and following LA treatment. Completion of the 5
year prospective observation period of the Pro(a)LiFe study
will add results to this assumption [21].
Our study confirmed the safety of LA treatment, as has
been documented in over 2,500,000 treatments worldwide.
With a total rate of just 3.3% in unexpected adverse events,
vascular access problems, and technical problems, and
a completion rate of 99.5% of all treatments, the safety of
the therapy matched what would be expected at a High
Volume Center. An inverse correlation between treatment
numbers and complications has been a major result of an
apheresis registry analysis [37]. Complications and prob-
lems accessing veins were 10e20 times more likely to
occur in centers with less apheresis experience than in High
Volume Centers. The high safety standard of lipoprotein
apheresis is pointed out when comparing our results to the
International Apheresis Registry, which reported side
effects or complications in 20.4% of 6787 apheresis
procedures performed on 1735 patients [38]. However,
comparability is limited as in this registry a wide variety of
apheresis methods and indications were pooled. The
tolerability of LA is closely linked with complication rates,
which is again reflected in treatment course adherence and
ultimately in clinical success. LA adherence per year with
99.5% was higher in this study than in any other published
therapeutic approach.
5. Conclusions
LA treatment of patients with high risk for CVD due to
LDL- and/or Lp(a) hyperlipoproteinemia is effective, safe,
and well tolerated. The number of cardiovascular events, at
least during a six-year period, was lowered by 80%. Risk
reduction was more effective in those patients with isolated
elevated Lp(a) than in those patients with severe hyper-
cholesterolemia. In addition to this, the risk reduction was
more effective in both treatment indications than one would
expect from the lowering of LDL-C and Lp(a) alone.
Epidemiological research and prospective studies
comparing LA and new Lp(a) lowering medication would
161
be useful in understanding the role of Lp(a) in cardiovas-
cular diseases.
Conflict of interest
F. Heigl received lecture fees from B. Braun, Melsun-
gen, Diamed, Cologne, Fresenius Medical Care, Bad
Homburg.
R. Klingel received financial support for clinical
research activities by grants from Asahi Kasei Medical,
Japan and Diamed, Germany.
Acknowledgments
This scientific work is dedicated to the patients and the
entire team of the Apheresis Competence Center at MVZ
Kempten-Allgaeu. We, the authors, as well as our patients
are grateful to all members of this highly qualified team of
nurses and medical and non-medical specialists who addi-
tional to their daily routines and with outstanding
commitment from a human as well professional perspective
made it possible to collect the huge data set for this clinical
investigation. Results showed minimal complication rates
and maximal patient compliance with lipoprotein apheresis
reflecting the exceptional level of patient care and organi-
zational standards. Representing the entire team we wish to
mention Ines Schulz-Merkel, Kerstin Rziha, Svende
Kiehstaller, Maria Rietzler, Waltraud Gast, and Ulrike
Sattler.
References
[1] Castelli WP. Epidemiology of coronary heart disease: the Framing-
ham study. Am J Med 1984;76(2A):4e12.
[2] Pedersen TR for the Scandinavian Simvastatin Survival Study (4S)
Group. Randomised trial of cholesterol lowering in 4444 patients
with coronary heart disease. Lancet 1994;344:1383e9.
[3] O’Keefe JH Jr, Cordain L, Harris WH, Moe RM, Vogel R. Optimal
low-density lipoprotein is 50 to 70 mg/dL: lower is better and
physiologically normal. J Am Coll Cardiol 2004;43(11):2142e6.
[4] Berg K. A new serum type system in man: the Lp system. Acta Pathol
Microbiol Scand 1963;59:369e82.
[5] Cremer P, Nagel D, Labrot B, Mann H, Muche R, Elster H, et al.
Lipoprotein Lp(a) as predictor of myocardial infarction in compar-
ison to fibrinogen, LDL cholesterol and other risk factors: results
from the prospective Goettingen Risk Incidence and Prevalence
Study (GRIPS). Eur J Clin Invest 1994;24(7):444e53.
[6] Emerging Risk Factors Collaboration. Lipoprotein(a) concentration
and the risk of coronary heart disease, stroke, and nonvascular
mortality. JAMA 2009;302:412e23.
[7] Kamstrup PR. Lipoprotein(a) and ischemic heart diseaseea causal
association? A review. Atherosclerosis 2010;211:15e23.
[8] Nordestgaard BG, Chapman MJ, Ray K, Borén J, Andreotti F,
Watts GF, et al; for the European Atherosclerosis Society Consensus
Panel. Lipoprotein(a) as a cardiovascular risk factor: current status.
Eur Heart J 2010;31:2844e53.
[9] Ariyo AA, Thach C, Tracy R, Cardiovascular Health Study Investi-
gators. Lp(a) lipoprotein, vascular disease, and mortality in the
elderly. N Engl J Med 2003;349(22):2108e15.
162 F. Heigl et al. / Atherosclerosis Supplements 18 (2015) 154e162
[10] Thompson GR, Lowenthal R, Myant NB. Plasma exchange in the
management of homozygous familial hypercholesterolemia. Lancet
1975;1:1208e11.
[11] Thompson GR, Myant NB, Kilpatrick D, Oakley CM, Raphael MJ,
Steiner RE. Assessment of long-term plasma exchange for familial
hypercholesterolemia. Br Heart J 1980;43:680e8.
[12] Agishi T, Kaneko I, Hasuo Y, Hayasaka Y, Sanaka T, Ota K, et al.
Double filtration plasmapheresis. Trans Am Soc Artif Intern Organs
1980;XXVI:406e11.
[13] Stoffel W, Borberg H, Greve V. Application of specific extracorporeal
removal of low density lipoprotein in familial hypercholesterolemia.
Lancet 1981;2:1005e7.
[14] Seidel D, Wieland H. A new method for selective measurement and
extracorporeal elimination of low density lipoproteins. (Article in
German). J Clin Chem Clin Biochem 1982;20:684e5.
[15] Yokoyama S, Hayashi R, Kikkawa T, Tani N, Takada S, Hatanaka K,
et al. Specific sorbent of apolipoprotein B-containing lipoproteins for
plasmapheresis. Characterization and experimental use in hypercho-
lesterolemic rabbits. Arteriosclerosis 1984;4(3):276e82.
[16] Bosch T, Schmidt B, Blumenstein M, Gurland HJ. Lipid apheresis by
hemoperfusion: in vitro efficacy and ex vivo biocompatibility of
a new low-density lipoprotein adsorber compatible with human
whole blood. Artif Organs 1993;17:640e52.
[17] Seidel D. H.E.L.P. Report 1994-10 Years of clinical experience.
MMV Medizin Verlag München; 19941e64.
[18] Mabuchi H, Koizumi J, Shimizu M, Kajinami K, Miyamoto S,
Ueda K, et al. Long-term efficacy of low-density lipoprotein apher-
esis on coronary heart disease in familial hypercholesterolemia.
Hokuriku-FH-LDL-Apheresis Study Group. Am J Cardiol 1998;
82(12):1489e95.
[19] Bosch T, Keller C. Clinical effects of direct adsorption of lipoprotein
apheresis: beyond cholesterol reduction. Ther Apher Dial 2003;7:
341e4.
[20] Jaeger BR, Richter Y, Nagel D, Heigl F, Vogt A, Roeseler E, et al.
Longitudinal cohort study on the effectiveness of lipid apheresis
treatment to reduce high lipoprotein(a) levels and prevent major
adverse coronary events. Nat Clin Pract Cardiovasc Med 2009;6(3):
229e39.
[21] Leebmann J, Roeseler E, Julius U, Heigl F, Spitthoever R,
Heutling D, et al; for the Pro(a)LiFe Study Group. Lipoprotein
apheresis in patients with maximally tolerated lipid-lowering therapy,
lipoprotein(a)-hyperlipoproteinemia, and progressive cardiovascular
disease: prospective observational multicenter study. Circulation
2013;128(24):2567e76.
[22] German Federal Ministry of Health. Publication of a decision of
Federal Joint Committee on apheresis for isolated Lp(a) elevation
with progressive cardiovascular disease. Trans Ger Fed Ministries
(BAnz) 2008;138. 3321 and most recent modification in January
2013 at Transactions of the German Federal Ministries (BAnz) AT
08.04.2013. (Articles in German).
[23] National association of statutory health insurance physicians (KBV).
Qual Rep 2013:47e8 [Article in German].
[24] Heigl F, Hettich R, Lotz N, Reeg H, Eder B, Steckholzer-Kroth K,
et al. Indication and implementation of lipidapheresis, rheopheresis
or immunoadsorption (Lessons learnt from Germany’s largest
apheresis center). Atheroscler Suppl 2009;10:137e41.
[25] Thompson GRHEART-UK LDL Apheresis Working Group.
Recommendations for the use of LDL apheresis. Atherosclerosis
2008;198:247e55.
[26] Thompson GR, Barbir M, Davies D, Dobral P, Gesinde M,
Livingston M, et al. Efficacy criteria and cholesterol targets for LDL
apheresis. Atherosclerosis 2010;208:317e21.
[27] Heigl F, Hettich R, Eder B, Arendt R. Lipoprotein apheresis standard
for apheresis competence centers e an updated synthesis and
amendment to pre-existing standards. Atheroscler Suppl 2013;14:
57e65.
[28] Schettler VScientific Board of GLAR and for the German Apheresis
Working Group. First data from the German lipoprotein apheresis
registry (GLAR). Atheroscler Suppl 2015;18:41e4.
[29] Kroon AA, van’t Hof MA, Demacker PN, Stalenhoef AF. The
rebound of lipoproteins after LDL-apheresis. Kinetics and estimation
of mean lipoprotein levels. Atherosclerosis 2000;152:519e26.
[30] Grundy SM, Cleeman JI, Merz CN, Brewer HB Jr, Clark LT,
Hunninghake DB, et al. Implications of recent clinical trials for the
national cholesterol education program adult treatment panel III
guidelines. Circulation 2004;110:227e39.
[31] von Dryander M, Fischer S, Passauer J, Müller G, Bornstein SR,
Julius U. Differences in the atherogenic risk of patients treated by
lipoprotein apheresis according to their lipid pattern. Atheroscler
Suppl 2013;14:39e44.
[32] Briel M, Ferreira-Gonzalez I, You JJ, Karanicolas PJ, Akl EA, Wu P,
et al. Association between change in high density lipoprotein
cholesterol and cardiovascular disease morbidity and mortality:
systematic review and meta-regression analysis. BMJ 2009;338:b92.
[33] Wang Y, Blessing F, Walli AK, Uberfuhr P, Fraunberger P, Seidel D.
Effects of heparin-mediated extracorporeal low-density lipoprotein
precipitation beyond lowering proatherogenic lipoproteinsereduction
of circulating proinflammatory and procoagulatory markers.
Atherosclerosis 2004;175(1):145e50.
[34] Kobayashi S, Moriya H, Maesato K, Okamoto K, Ohtake T. LDL-
apheresis improves peripheral arterial occlusive disease with an
implication for anti-inflammatory effects. J Clin Apher 2005;20(4):
239e43.
[35] Moriarty PM, Gibson CA, Shih J, Matias MS. C-reactive protein and
other markers of inflammation among patients undergoing HELP
LDL apheresis. Atherosclerosis 2001;158(2):495e8.
[36] Wieland E, Schettler V, Armstrong VW. Highly effective reduction of
C-reactive protein in patients with coronary heart disease by extra-
corporeal low density lipoprotein apheresis. Atherosclerosis 2002;
162(1):187e91.
[37] Klingel R, Fassbender C, Heibges A, Koch F, Nasemann J,
Engelmann K, et al. RheoNet registry analysis of rheopheresis for
microcirculatory disorders with a focus on age-related macular
degeneration. Ther Apher Dial 2010;14:276e86.
[38] Malchesky PS, Koo AP, Skibinski CI, Hadsell AT, Rybicki LA.
Apheresis technologies and clinical applications: the 2007 interna-
tional apheresis registry. Ther Apher Dial 2010;14:52e73.