Comparison of Different Low Density LipoproteinApheresis Machines on Brain Natriuretic Peptide Levels in

Patients With Familial Hypercholesterolemia

Patrick Moriarty, Rachel Sosland, Cheryl Gibson, and John Belmont

Department of Internal Medicine, University of Kansas Medical Center, Kansas City, KS, USA

Abstract: B-type natriuretic peptide (BNP) is a hormonereleased from cardiac ventricles during episodes ofhemodynamic overload. Low density lipoprotein (LDL)apheresis, a procedure for patients with familial hypercho-lesterolemia (FH) and coronary artery disease (CAD),lowers plasma cholesterol and immediately reduces bloodviscosity and coronary vascular resistance while improvingmyocardial blood flow and microvascular perfusion. Previ-ous studies have demonstrated the ability of LDL apheresisto reduce BNP chronically. We undertook this study toevaluate the difference in reduction of BNP levels follow-ing a single treatment with two dissimilar LDL apheresisdevices. We conducted a prospective trial involving 27patients (19 F; age = 59 � 9 years) with FH who received atleast 6 months of bi-weekly LDL apheresis therapy witheither the Secura heparin extracorporeal LDL precipita-tion (HELP) system (N = 17 patients, B. Braun, Inc., Mel-

sungen, Germany) or the Liposorber LA-15 dextran sulfateabsorber (DSA) system (N = 10 patients, Kaneka, Inc.,Osaka, Japan). We measured BNP levels immediatelybefore and after one treatment of LDL apheresis. Follow-ing LDL apheresis, BNP levels were reduced by an averageof 40 � 17% (P < 0.001). Despite treating equal amountsof plasma, the HELP system reduced BNP (45 � 18%)significantly more than the DSA system (31 � 11%,P = 0.031). In conclusion, LDL apheresis therapy, possiblythrough its immediate improvement of vascular flowand/or removal of the peptide from plasma, results in asignificant reduction of BNP levels. The increased reduc-tion of BNP by HELP may result from its superior acutealterations of rheological markers. Key Words: BNP,Familial hypercholesterolemia, Hemorheology, Heartfailure, LDL apheresis, Viscosity.tap_692 74..78

Brain natriuretic peptide (BNP) and the inactiveN-terminal segment of BNP (N-BNP) are peptidehormones released into the vasculature in responseto increased cardiac wall stress, as seen with con-gestive heart failure (CHF) (1). Their physiologicalroles include vasodilatation, diuresis, natriuresis andhypotension (2). Although BNP/N-BNP have dem-onstrated their significance in diagnosing and assist-ing in the therapy of heart failure, it has becomeapparent that many patients with elevated BNP havenormal ventricular function, and raised levels maysignify more cardiac structural disease than simplyCHF (3). The majority of natriuretic peptides areremoved from the circulation by way of enzyme deg-

radation, endopeptidases, clearance from the kidneyand receptor-mediated endocytosis (4) with a half-life of roughly 30 min (5).

Familial hypercholesterolemia (FH) is an autoso-mal dominant disorder characterized by elevatedplasma cholesterol and premature coronary heartdisease (CHD) in the absence of other risk factors(6). For FH patients resistant to pharmacotherapy,LDL apheresis has proven effective for removingcholesterol and preventing future cardiovascularevents (7). The procedure, performed every1–2 weeks, can reduce low density lipoprotein choles-terol (LDL-C) by 60 to 80% and lower other proat-herogenic plasma proteins, such as fibrinogen, CRPand Lp-PLA2. Immediate vascular changes followingLDL apheresis include improvement of blood rheol-ogy, endothelial function, myocardial blood flowand microvascular perfusion (8–10). In this study weexamined the acute reduction of BNP in FH patients

Received October 2008; revised January 2009.Address correspondence and reprint requests to Dr Patrick

Moriarty, 3901 Rainbow Boulevard., Kansas City, KS 66160, USA.Email: pmoriart@kumc.edu

Therapeutic Apheresis and Dialysis 14(1):74–78doi: 10.1111/j.1744-9987.2009.00692.x© 2009 The AuthorsJournal compilation © 2009 International Society for Apheresis

74

with dyslipidemia and not necessarily CHF, receivingLDL apheresis (dextran sulfate absorber [DSA] andheparin extracorporeal LDL precipitation [HELP]).Based on the rapid enhancement of vascular hemo-dynamics subsequent to LDL apheresis we hypoth-esized that BNP levels would be significantly reducedfollowing LDL apheresis. We also hypothesized thatthe HELP system would lower BNP levels more thanthe DSA system due to its greater acute effects onrheological markers.

METHODS

This is a prospective, single center study. The Uni-versity of Kansas Institutional Review Boardreviewed and approved the protocol. All enrolledpatients gave written informed consent.

PatientsWe selected 27 FH patients (see Table 1 for base-

line characteristics) from the Kansas University Ath-erosclerosis and LDL Apheresis Center. Selectionwas based on who had been receiving bi-weekly LDLapheresis for at least 6 months. All patients qualifiedfor LDL apheresis based on the US Food and DrugAdministration criteria which requires patients to berefractory to cholesterol reducing medications withuncontrolled hypercholesterolemia (LDL >200 mg/dL) and CHD or hypercholesterolemia (>300 mg/dL) without CHD.

Device description and procedureSince 1996 the Food and Drug Administration has

approved two LDL apheresis devices for clinical usein the United States: (i) The dextran sulfate absorber(DSA) Liposorber LA-15 system (Kaneka, Osaka,Japan) (11) separates blood from the heparinizedplasma with a polysulfone plasmafilter. The plasma isthen exposed to a dextran sulfate cellulose column,which absorbs apo-B lipoproteins. The machine con-tains two identical columns. After 500 mL of plasma

has passed through the first column, a 4.1% sodiumchloride solution rinses and regenerates the column.During the rinsing process plasma flow is redirectedto the other column; and (ii) the heparin extracor-poreal LDL precipitation (HELP) Secura system(Braun, Melsungen, Germany) (12) separates bloodcells through a 0.2 micron plasma filter at flow ratesof 70 mL/min to continuously remove plasma at 20 to30 mL/min. The plasma is then mixed 1 : 1 with a0.3 M acetate buffer (pH 4.8) solution containingheparin at a concentration of 100 U/mL. Precipita-tion of heparin and LDL-C occurs when the plasmabuffer solution reaches an approximate pH of 5.2and the precipitate is then separated by a secondfilter. The residual heparin in the LDL-free plasmais adsorbed by a diethyaminoethyl cellulose filter.Physiological pH of the plasma and removal of excessfluid is achieved by dialysis and ultrafiltration.

Patients undergoing apheresis have both antecu-bital veins placed with a 16–18 G needle while lyingin a hospital chair or bed. A normal procedure treatsabout 3–5 L of plasma over a 1–2 h period, witheither HELP or DSA, and at any time during theprocedure only 300 to 400 mL is found to be extra-corporeal. Venous shunts or fistulas are sometimesrequired for patients with poor vascular access.Depending on how much plasma is treated, LDL-Ccan be immediately reduced by 60%-80% (13).Treat-ments usually take place every two weeks.

Laboratory measurementsTotal cholesterol (TC), triglycerides, and high-

density lipoprotein cholesterol (HDL) were deter-mined by enzymatic colorimetric test kits (VITROSchemistry systems, Ortho-Clinical Diagnostics, Inc.,Raritan, NJ, USA). B-type natriuretic peptide wasmeasured immediately before and after each treat-ment using the Clinical Laboratories ImprovementAct (CLIA) approved immunoassay test (TriageBNP; Biosite Diagnostics, San Diego, CA, USA) (14).Levels were measured immediately before and aftera single treatment of LDL apheresis. Biosite detectsthe biologically active forms of BNP, but does notmeasure the metabolically inert BNP (2).

Statistical analysesData are reported as mean � standard deviation

unless otherwise noted. Statistical analysis includedthe Mann-Whitney test for pre- and post-apheresischanges in lipid profiles and BNP levels. Differencesalso were assessed between the HELP and DSA aph-eresis procedures and their effects on changes in lipidprofiles and BNP levels. Spearman’s rank correlationcoefficients were used to measure the association

TABLE 1. Baseline patient characteristics (N = 27)

Variable Value

Age (years), mean � SD 59.2 � 9.4Sex (male, %) 8 (30)EF%, mean � SD 56.8 � 10.5*On statins (N, %) 10 (37)On fish oil/omega3 (N, %) 14 (52)Diabetes Mellitus (N, %) 8 (30)Hypertension (N, %) 21 (78)Coronary Heart Disease (N, %) 18 (68)

*Data available only on 19 patients; EF, ejection fraction; SD,standard deviation.

LDL Apheresis and BNP Levels 75

© 2009 The AuthorsJournal compilation © 2009 International Society for Apheresis Ther Apher Dial, Vol. 14, No. 1, 2010

between changes in BNP and changes in TC, triglyc-erides, HDL-cholesterol and LDL-cholesterol levels.Statistical analyses were performed with SPSS version14 (SPSS Inc., Chicago, IL, USA). A two-tailed a of0.05 was used to denote statistical significance.

RESULTS

B-type natriuretic peptide levels were significantlyreduced by 40% (P < 0.001) (see Table 2), andLDL-C levels were lowered by 59%.The % change inBNP correlated with % change in both TC (r = 0.342,P = 0.081) and HDL (r = 0.474, P = 0.013) but notLDL-C.

The HELP and DSA machines did not differ sig-nificantly on pre-BNP or post-BNP measurements,but the difference between their mean % changeswas significant (P = 0.031), and the change in rawBNP from pre to post-apheresis was significantfor both machines. For HELP the change was122 � 142 pg/mL to 63 � 70 pg/mL (P = 0.007). ForDSA the change was 140 � 160 pg/mL to94 � 109 pg/mL (P = 0.021) (Table 3).

DISCUSSION

Low density lipoprotein apheresis immediatelyreduces plasma levels of BNP by 40%. The mecha-nism by which BNP is lowered following LDL

apheresis is not fully understood. We hypothesize thedrop in BNP levels after LDL apheresis is a conse-quence of decreased hemodynamic stimulation ofcardiac tissue resulting in a diminished secretion ofthe peptide hormone and the direct removal of BNPfrom the treated plasma.

B-type natriuretic peptide release is stimulated byhemodynamic stress, such as wall stretch, ventriculardilation, and/or increased pressure from fluid over-load. A large portion of the external work performedby the heart is to eject blood into the pulmonaryartery and aorta. This work is termed stroke work,and is calculated as volume of blood ejected duringthe cardiac cycle or stroke volume (SV) multiplied bythe pressure (P) at which the blood is ejected: strokework = P ¥ SV. Pressure (P) is equal to flow (F) mul-tiplied by vascular resistance (R): P = F ¥ R. Flow (F)is equal to pressure (P) multiplied by the vesselradius (r) to the fourth power divided by blood vis-cosity (h) multiplied by the vessel length (L):F = P ¥ r4/h ¥ L. Resistance (R) is equal to viscosity(h) multiplied by the length of the vessel (L) dividedby the radius (r) to the fourth power: R = h ¥ L/r4

(15). Blood viscosity is the frictional force that devel-ops between adjacent layers of blood and is alteredby shear forces (rate and stress), hematocrit, redblood cell (RBC) aggregation, RBC deformabilityand plasma viscosity (16). A reduction of blood vis-cosity will result in a decrease of vascular resistanceand improvement of cardiac blood flow. An increasein systemic vascular resistance, as seen with CHF, willstimulate cardiac muscle to secrete BNP.

Low density lipoprotein apheresis immediatelyimproves endothelial function, myocardial bloodflow, and lowers coronary resistance. The vascularrecovery observed following LDL apheresis may beassociated with the positive changes of vasoreactivesubstances, such as nitric oxide (17), bradykinin(18), PGI2 (19), insulin-like growth factor (20) andendothelin-1 (21). Through its ability to improveRBC aggregation, RBC deformability and plasmaviscosity, LDL apheresis decreases blood viscosity

TABLE 2. Comparison of B-type natriuretic peptide and cholesterol levels immediately before and after low densitylipoprotein apheresis (mean � standard error)

Parameter Pre-apheresis Post-apheresis % ChangeP-value

(2-tailed)

BNP (pg/mL) 129 � 28 74 � 17 -40 � 3.2 <0.001LDL-cholesterol (mg/dL) 203 � 13 79 � 6 -59 � 1.8 <0.001Total cholesterol (mg/dL) 265 � 14 129 � 7 -50 � 1.7 <0.001Triglycerides (mg/dL) 177 � 17 88 � 13 -53 � 2.9 <0.001HDL-cholesterol (mg/dL) 41 � 2 35 � 2 -15 � 1.7 <0.001

HDL, high density lipoprotein.

TABLE 3. Pre-apheresis and post-apheresis values forB-type natriuretic peptide (mean � standard deviation)

with two different apheresis procedures

Measure HELP DSA P-value

Pre-apheresis BNP(pg/mL)

122 � 142 140 � 160 0.775

Post-apheresis BNP(pg/mL)

63 � 70 94 � 109 0.361

% change 45 � 17.6 31 � 11.2 0.031

DSA, dextran sulfate absorber; HELP, heparin extracorporealLDL precipitation.

P Moriarty et al.76

© 2009 The AuthorsJournal compilation © 2009 International Society for ApheresisTher Apher Dial, Vol. 14, No. 1, 2010

(10% to 15%) for at least one week and improvesperipheral vascular resistance for 3 weeks (22). Thehalf-life of human BNP is approximately 20 min(23) and plasma reductions can occur within 60 minafter improvement of hemodynamic function inpatients with CHF (24). The observed improve-ments in vascular hemodynamics and loweredplasma peptide levels following LDL apheresis mayinduce a rapid inactivation of BNP gene expressionin the heart.

Direct removal is another possible mechanism bywhich LDL apheresis reduces plasma BNP. BothLDL apheresis devices (i.e. HELP and DSA) usedin this study remove LDL and other lipoproteinsthrough an interaction between polyanions and pro-teins. The negatively charged dextran sulfate filter(DSA) and the low pH (5.12) acetate buffer withpolyanion heparin (HELP) form non-soluble precipi-tates with positively charged amino acid domains inapolipoprotein B while negatively charged plasmaproteins, such as HDL, are spared. In plasma, BNPhas an isoelectric charge of 10.9 (25). The positivecharge of the peptide may predispose its removalduring LDL apheresis.

CONCLUSION

In our analyses, we found HELP lowered BNPlevels significantly more than DSA.We postulate thatthe greater improvements gained through the use ofthe HELP system results from its increase effect onvascular hemodynamics. The HELP system acutelyreduces more plasma fibrinogen than the DSAsystem. This protein (fibrinogen) is associated withincreasing red blood cell (RBC) aggregation andplasma viscosity with no significant effect on RBCdeformability. In a previously published comparisontrial between HELP and DSA, the HELP system wassuperior to the DSA system in reducing blood viscos-ity (15% vs. 5%) and RBC aggregation (50% vs.30%) while both machines equally improved RBCdeformability, the authors suggested the differencebetween the two machines effect on rheology was aresult of their difference in reducing plasma fibrino-gen levels (62% vs. 11%) (26).

Previous research has demonstrated long termLDL apheresis therapy chronically reduces BNP(17%) (27), C-reactive protein (50%) (28,29), andLp-PLA2 (20%) (30) irrespective of changes toLDL-C. The acute and chronic decrease in BNPlevels and other plasma proteins support the hypoth-esis that LDL apheresis may improve vascular func-tion independent of changes in lipid levels.

LimitationsThere are several limitations to this study. We are

examining a limited number of FH patients withdiverse baseline characteristics, including hyperten-sion, diabetes and coronary artery disease. B-typenatriuretic peptide levels at baseline were notelevated to a clinically significant level; however,BNP was significantly reduced by 40%. Our hypoth-esis of BNP levels reduced through changes in hemo-dynamics was not supported by simultaneouslymeasuring rheology markers during the study but bypreviously published data comparing the two devices.Additional studies are needed to evaluate the impactof LDL apheresis mediated changes on BNP and itspotential application for patients with congestiveheart failure and elevated BNP levels.

REFERENCES

1. Karthikeyan VJ, Lip GY. N-terminal pro-brain natriureticpeptide and coronary artery disease. Eur J Clin Invest 2007;37:1–7.

2. Dries DL. Relevance of molecular forms of brain natriureticpeptide for natriuretic peptide research. Hypertension 2007;49:971–3.

3. Omland T, Sabatine MS, Jablonski KA et al. Prognostic valueof B-Type natriuretic peptides in patients with stable coronaryartery disease: the PEACE Trial. J Am Coll Cardiol 2007;50:205–14.

4. Vanderheyden M, Bartunek J, Goethals M. Brain and othernatriuretic peptides: molecular aspects. Eur J Heart Fail 2004;6:261–8.

5. Kroll MH, Twomey PJ, Srisawasdi P. Using the single-compartment ratio model to calculate half-life, NT-proBNP asan example. Clin Chim Acta 2007;380:197–202.

6. Hobbs HH, Brown MS, Goldstein JL. Molecular genetics ofthe LDL receptor gene in familial hypercholesterolemia. HumMutat 1992;1:445–66.

7. Mabuchi H, Koizumi J, Shimizu M et al. Long-term efficacy oflow-density lipoprotein apheresis on coronary heart disease infamilial hypercholesterolemia. Hokuriku-FH-LDL-ApheresisStudy Group. Am J Cardiol 1998;82:1489–95.

8. Moriarty PM, Gibson CA, Kensey KR, Hogenauer W. Effectof low-density lipoprotein cholesterol apheresis on blood vis-cosity. Am J Cardiol 2004;93:1044–6.

9. Igarashi K, Tsuji M, Nishimura M, Horimoto M. Improvementof endothelium-dependent coronary vasodilation after a singleLDL apheresis in patients with hypercholesterolemia. J ClinApher 2004;19:11–16.

10. Mellwig KP, Baller D, Gleichmann U et al. Improvement ofcoronary vasodilatation capacity through single LDL apher-esis. Atherosclerosis 1998;139:173–8.

11. Mabuchi H, Michishita I, Takeda M et al. A new low densitylipoprotein apheresis system using two dextran sulfate cellu-lose columns in an automated column regenerating unit (LDLcontinuous apheresis). Atherosclerosis 1987;68:19–25.

12. Wieland H,Seidel D.A simple specific method for precipitationof low density lipoproteins. J Lipid Res 1983;24:904–9.

13. Klingel R, Fassbender T, Fassbender C, Gohlen B. From mem-brane differential filtration to lipidfiltration: technologicalprogress in low-density lipoprotein apheresis. Ther Apher Dial2003;7:350–8.

14. Fischer Y, Filzmaier K, Stiegler H et al. Evaluation of a new,rapid bedside test for quantitative determination of B-typenatriuretic peptide. Clin Chem 2001;47:591–4.

LDL Apheresis and BNP Levels 77

© 2009 The AuthorsJournal compilation © 2009 International Society for Apheresis Ther Apher Dial, Vol. 14, No. 1, 2010

15. Cheruvu PK, Finn AV, Gardner C et al. Frequency and distri-bution of thin-cap fibroatheroma and ruptured plaques inhuman coronary arteries: a pathologic study. J Am CollCardiol 2007;50:940–9.

16. Kaul S, Jayaweera AR. Determinants of microvascular flow.Eur Heart J 2006;27:2272–4.

17. Tasaki H, Tasaki H. Low-density lipoprotein apheresis in theprevention of recurrent coronary heart disease: a review. TherApher Dial 2003;7:408–12.

18. Krieter DH, Steinke J, Kerkhoff M et al. Contact activation inlow-density lipoprotein apheresis systems. Artif Organs 2005;29:47–52.

19. Mii S, Mori A, Sakata H, Nakayama M, Tsuruta H. LDL aph-eresis for arteriosclerosis obliterans with occluded bypassgraft: change in prostacyclin and effect on ischemic symptoms.Angiology 1998;49:175–80.

20. Kobayashi S, Moriya H, Negishi K, Maesato K, Ohtake T.LDL-apheresis up-regulates VEGF and IGF-I in patients withischemic limb (see comment). J Clin Apheresis 2003;18:115–19.

21. Nakamura T, Ushiyama C, Osada S et al. Effect of low-densitylipoprotein apheresis on plasma endothelin-1 levels in diabetichemodialysis patients with arteriosclerosis obliterans. J DiabetComplications 2003;17:349–54.

22. Rubba P, Iannuzzi A, Postiglione A et al. Hemodynamicchanges in the peripheral circulation after repeat low densitylipoprotein apheresis in familial hypercholesterolemia. Circu-lation 1990;81:610–16.

23. Holmes SJ, Espiner EA, Richards AM, Yandle TG, FramptonC. Renal, endocrine, and hemodynamic effects of human brain

natriuretic peptide in normal man. J Clin Endocrinol Metab1993;76:91–6.

24. Gegenhuber A, Mueller T, Firlinger F, Lenz K, Poelz W, Halt-mayer M. Time course of B-type natriuretic peptide (BNP)and N-terminal proBNP changes in patients with decompen-sated heart failure. Clin Chem 2004;50:454–6.

25. Clemens LE, Almirez RG, Baudouin KA, Mischak RP, Gross-bard EB, Protter AA. Pharmacokinetics and biological actionsof subcutaneously administered human brain natriureticpeptide. J Pharmacol Exp Ther 1998;287:67–71.

26. Grutzmacher P, Landgraf H, Esser R et al. In vivo rheologiceffects of lipid apheresis techniques: comparison of dextransulfate LDL adsorption and heparin induced LDL precipita-tion. ASAIO Trans 1990;36:M327–30.

27. Jaeger BR. The HELP system for the treatment of athero-thrombotic disorders: a review. Ther Apher Dial 2003;7:391–6.

28. Otto C, Geiss HC, Empen K, Parhofer KG. Long-term reduc-tion of C-reactive protein concentration by regular LDL aph-eresis. Atherosclerosis 2004;174:151–6.

29. Kobayashi J, Katsube S, Shimoda M et al. Single LDL apher-esis improves serum remnant-like particle-cholesterol,C-reactive protein, and malondialdehyde-modified-low-density lipoprotein concentrations in Japanese hypercholester-olemic subjects. Clin Chim Acta 2002;321:107–12.

30. Moriarty PM, Gibson CA. Effect of low-density lipoproteinapheresis on lipoprotein-associated phospholipase A2. Am JCardiol 2005;95:1246–7.

P Moriarty et al.78

© 2009 The AuthorsJournal compilation © 2009 International Society for ApheresisTher Apher Dial, Vol. 14, No. 1, 2010