Chelation Therapy

Home , Chelation therapy

รายงานการทบทวนวรรณกรรม

เรื่อง

Chelation therapy

สําหรับโรคหลอดเลือดหัวใจ

โดย

นายแพทยอรรถสิทธิ์ ศรีสุบัติ

สถาบันวจิ ัยและประเมินเทคโนโลยทางการแพทยี  กรมการแพทย กระทรวงสาธารณสุข พ.ศ.2553 ก

คํานํา

การรักษาภาวะที่มีโลหะหนักในรางกายโดยวิธีการ Chelation มีการใชมานานในผูปวย โรคธาลัสซีเมีย (thalassemia) ที่ไดรับเลือดจะมีธาตุเหล็กเกินในรางกาย ผูปวยจะไดรับการรักษาดวยการให deferoxamine ซึ่งเปนยามาตรฐานในการรักษา ดวยคุณสมบัติของวิธี Chelation therapy ในการจับโลหะ หนักและขับออกจากรางกายได จึงเปนที่มาของแนวคิดที่วาวิธี Chelation therapy ดวยสาร EDTA (ethylenediamine tetraacetic acid) สามารถกําจัดโลหะหนักที่บริเวณผนังหลอดเลือดหัวใจไดเชนกัน ทําให ลดการเกิด atherosclerotic plaque ที่ผนังหลอดเลือดหัวใจลงได ดวยเหตุนี้ การรักษาดวยวิธีการแพทย ทางเลือกจึงมีการนํา Chelation therapy มาใชในการรักษาโรคหลอดเลือดหัวใจในปจจุบันอยางแพรหลาย อยางไรก็ตาม ยังมีขอโตเถียงและความคิดเห็นที่แตกตางกันของผูเชี่ยวชาญแขนงตางๆ ในประเด็นของ ประสิทธิผลของการใช Chelation therapy ในการรักษาโรคหลอดเลือดหัวใจ รวมถึงผลขางเคียงที่อาจเกิด ขึ้นกับผูปวย

ดังนั้น กรมการแพทย โดย สถาบันวิจัยและประเมินเทคโนโลยีทางการแพทย จึงไดดําเนินการ ทบทวนวรรณกรรมที่มีหลักฐานนาเชื่อถือที่มีอยูในปจจุบันบนพื้นฐานของขอมูลที่สามารถสืบคนและเขาถงึ ได เพื่อใหทราบถึงประสิทธิผลของ Chelation therapy ในการรักษาโรคหลอดเลือดหัวใจ

ข

บทคัดยอ

ปจจุบันมีการนํา Chelation therapy มาใชเปนการแพทยทางเลือกของการรักษา หรือใชเพื่อ ประกอบการรักษาภาวะ atherosclerosis และลดอาการของโรคหลอดเลือดหัวใจ ซึ่งเปนวิธีที่ทําใหมีเลือด ไหลเวียนในหลอดเลือดหัวใจที่แข็งโดยไมตองผาตัด จึงสามารถปองกันอาการตางๆ ของโรคหลอดเลือด หัวใจได อาทิ อาการปวดเคนหัวใจ (angina pectoris) หรือ กลามเนื้อหัวใจตาย (myocardial infarction) เปนตน อยางไรก็ตาม การใช Chelation therapy ในปจจุบันเพื่อรักษา atherosclerosis และโรคหลอดเลือด หัวใจยังมีขอขัดแยงในดานประสิทธิผลและความปลอดภัยในกลุมผูเชี่ยวชาญแขนงตางๆ การศึกษานี้จึงมี วัตถุประสงคในการหาประสิทธิผลและผลขางเคียงจาก Chelation therapy ที่นํามารักษาโรคหลอดเลือด หัวใจ โดยการทบทวนวรรณกรรมจากการสืบคนเอกสารวิชาการที่เปนภาษาอังกฤษใน PubMed และ The Cochrane Library โดยผลการสืบคนได 777 เรื่อง (PubMed 615 และ The Cochrane Library 162 เรื่อง) ภายหลังจากการคัดเขาและออกตามเกณฑที่กําหนดพบวาเอกสารที่ไดเปนการทบทวนอยางเปนระบบ (systematic review) มีจํานวน 3 เรื่อง การทบทวนเชิงบรรยาย (narrative review) จํานวน 2 เรื่อง และการวิจัย คลินิกที่เปน randomized controlled trials จํานวน 2 เรื่อง ผลการศึกษาพบวา 6 การศึกษารายงานวาผลของ การใช Chelation therapy สําหรับโรคหลอดเลือดหัวใจไมแตกตางจากกลุมเปรียบเทียบ ในขณะที่มีรายงาน ถึงผลอันไมพึงประสงคของ Chelation therapy จาก EDTA (ethylenediamine tetraacetic acid) ที่พบใน การศึกษาเหลานี้ ไดแก พิษตอไต ชัก แคลเซียมในเลือดต่ํา น้ําตาลในเลือดต่ํา เปนตน ในขณะที่ การศึกษาที่เปนการทบทวนและเผยแพรในวารสารดานการแพทยทางเลือกเพียงการศึกษาเดียวที่สนับสนุน Chelation therapy ที่ใช EDTA มีประโยชนในผูปวยที่ใสสายสวนหลอดเลือดชนิดเคลือบยา เมื่อเปรียบเทียบ กับการใหยา clopidrogrel รวมกับ aspirin และไมพบผลขางเคียงที่สําคัญจาก Chelation therapy ดังนั้น ดวย ขอมูลที่มีอยูในปจจุบันจึงยังไมสามารถระบุถึงประสิทธิผลของ Chelation therapy ในการรักษาโรคหลอด เลือดหัวใจได หากแตตองระมัดระวังผลขางเคียงที่อาจเกิดขึ้นและเปนอันตรายตอผูปวยไวดวย การศึกษาที่ เปน randomized controlled trial ดวยระเบียบวิธีวิจัยที่ถูกตอง เปนไปตามหลักการปฏิบัติการวิจัยทางคลินิก ที่ดี (good clinical practice) มีจํานวนผูเขารวมโครงการเพียงพอ โปรงใส ปราศจากอคติ ตรวจสอบไดทุก ขั้นตอนการวิจัย และผูวิจัยไมมีประโยชนทับซอนจะสามารถแสดงใหเห็นถึงประสิทธิผลของ Chelation therapy สําหรับโรคหลอดเลือดหัวใจตอไปไดอยางชัดเจน

ค

กิตติกรรมประกาศ

รายงานการทบทวนวรรณกรรมครั้งนี้สําเร็จลุลวงตามวัตถุประสงคทุกประการ เนื่องจากไดรับ ความรวมมือและการสนับสนุนการดําเนินการเปนอยางดีจากผูมีสวนเกี่ยวของทุกฝาย รวมถึง ขอขอบคุณ หองสมุดกรมการแพทยที่ไดใหความชวยเหลือในการสืบคนเอกสารวิชาการจากเครือขายแหลงขอมูลตางๆ จนการดําเนนการแลิ วเสร็จ ผูวิจัยจึงขอขอบคุณมา ณ โอกาสนี้ ง

สารบัญ

หนา

คํานํา ก บทคัดยอ ข กิตติกรรมประกาศ ค สารบัญ ง สารบัญตาราง จ สารบัญแผนภมู ิ จ บทนํา 1 วัตถุประสงค 2 วิธีดําเนินการ 2 เกณฑการพิจารณาคัดเลอกเอกสารวื ิจัยสําหรับการทบทวน 3 กลุมประชากรศึกษา 3 การวัดผลลัพธ 3 วิธีการสืบคนขอมูล 3 ประโยชนที่คาดวาจะไดรับ 5 ขอจํากัดการศึกษา 5 ผลการดําเนินการ 5 สรุปผล อภิปราย และขอเสนอแนะ 9 เอกสารอางอิง 11 ภาคผนวก Chelation Therapy for Cardiovascular Disease Chelation therapy for coronary heart disease: An overview of all clinical investigations Chelation Therapy for Ischemic Heart Disease Effect of Chelation Therapy on Endothelial Function in Patients With Coronary Artery Disease: PATCH Substudy Chelation therapy for atherosclerotic cardiovascular disease (The Cochrane Review) EDTA chelation therapy for cardiovascular disease: a systematic Review Should EDTA Chelation Therapy be Used Instead of Long-term Clopidogrel plus Aspirin to Treat Patients at Risk from Drug-Eluting Stents?

จ

สารบัญตาราง

หนา

ตารางที่ 1 คําสืบคนและผลทไดี่ จาก PubMed 4 ตารางที่ 2 คําสืบคนและผลทไดี่ จาก The Cochrane Library 4 ตารางที่ 3 ประเภทและจานวนเอกสารวํ ิชาการทผี่ านการคัดเลือกเพอทบทวนื่ 6

สารบัญแผนภูมิ

หนา

แผนภูมิที่ 1 ผลการสืบคนขอมูล 5 บทนํา

Chelation therapy1-5เปนการกําจัดโลหะหนักโดยการใชสารบางชนิดเพื่อเขาไปจับกับโลหะหนัก แลวขับออกจากรางกาย โลหะหนักสวนใหญ ไดแก ตะกั่ว สารหนู ปรอท เหล็ก เปนตน สารที่ใชในการ กําจัดโลหะหนักมีการนํามาใชทางการแพทยตั้งแตสมัยสงครามโลกครั้งที่ 1 โดยเริ่มแรกใช dimercaprol หรือ รูจักกันในชื่อ British Anti-Lewisite (BAL) เปนยาแกพิษ (antidote) ของแกสอารเซ-นิกซึ่งเปนแกสพิษ โดย BAL จะไปจับกับอารเซนิกแลวเปลี่ยนรูปเปนสารประกอบที่ละลายน้ําได ผานเขาสูกระแสโลหิตแลว ไปขับออกที่ไตและตับ อยางไรก็ตาม BAL มีผลขางเคียงรายแรงตอรางกายเปนอยางมาก ตอมา Chelation therapy ถูกนํามาใชในประเทศอังกฤษภายหลังสงครามโลกครั้งที่ 2 เพื่อรักษาผูที่ไดรับพิษสารหนู ตะกั่ว และโลหะหนักอื่นๆ การให Chelating agent สามารถใหไดทางการรับประทาน ใหทางหลอดเลือด และทาง กลามเนื้อได สารที่นิยมใชในการทํา Chelation คือ EDTA (ethylenediamine tetraacetic acid)6-7 ซึ่งเปนกรด อะ มิโนสังเคราะห โดยทั่วไป EDTA ถูกนํามาใชเปนสารปองกันไมใหเลือดแข็งตัวในหองปฏิบัติการ โดยการ ยับยั้งการรวมตัวกันของเกล็ดเลือด Chelation therapy ไดใช EDTA โดยการใหทางหลอดเลือดดํารวมกับ การรับประทานวิตามินและเกลือแร EDTA เปนสารประกอบที่ละลายน้ําได ทั้งนี้ EDTA จะชวยในการ กําจัดตะกั่ว ปรอท แคดเมียม และเหล็กออกจากระบบไหลเวียนโลหิต ซึ่งเหล็กและแคดเมียมมีความสัมพันธ กับความดันโลหิตสูง ในขณะที่เหล็กปริมาณสูงมีผลตอโรคหัวใจ องคการอาหารและยาของสหรัฐอเมริกา (U.S. Food and Drug Administration, FDA) อนุญาตใหใช EDTA ในการรักษาพิษจากโลหะหนักบางชนิด อาทิ inorganic lead และใชรักษาพิษจาก digitalis เรื่อยมาจนกระทั่งมี antibody fragment มาใชรักษาพิษจาก digitalis ทดแทน อยางไรก็ตาม FDA ยังไมอนุญาตใหใช Chelation therapy ในการรักษาโรคหลอดเลือด หัวใจโคโรนารีย8 จากคุณสมบัติของ EDTA ในการใชจับสารโลหะหนักและขับออกจากเลือดได นักวิทยาศาสตรจึง ไดตั้งสมมติฐานวาหลอดเลือดแดงที่แข็งนาจะใหนิ่มลงได โดยการให EDTA เพื่อไปลด atherosclerotic plaque ที่ผนังหลอดเลือดหัวใจ9 เนื่องจาก EDTA สามารถดึงแคลเซียมออกจากผนังหลอดเลือดแดงได และ ในทศวรรษที่ 1950 มีผูสังเกตเห็นวา ผูปวยที่เปนโรคหลอดเลือดหัวใจแข็งและไดรับพิษจากตะกั่ว เมื่อได EDTA แลวพบวามีอาการของโรคหลอดเลือดแดงหัวใจดีขึ้นดวย10 ในป ค.ศ.1956 มีรายงานผูที่ปวยเปนโรค หลอดเลือดสวนปลายอุดตัน (occlusive peripheral vascular disease) รายงานวารูสึกดีขึ้นภายหลังจากไดรับ EDTA ตอมามีรายงานผูปวยอีกหลายรายเกี่ยวกับประโยชนของ EDTA ในการรักษาผูปวยโรคหลอดเลือด หัวใจแข็ง11 อยางไรก็ตาม ยังไมมีการประเมินผลขางเคียงที่รุนแรงในการศึกษาขนาดใหญที่มีการสุมกลุมศึกษา และกลุมเปรียบเทียบ แตมีรายงานพิษที่เกิดขึ้นจากการรายงานผูปวย พบวา EDTA มีพิษตอไต พิษอื่นๆ ที่ 2

อาจเกิดขึ้นจาก EDTA ไดแก แคลเซียมในเลือดต่ําทําใหเกิดการชักเกร็ง ซึ่งทําใหเสียชีวิตได ภาวะน้ําเกินใน รางกายทําใหเกิดหัวใจลมเหลว คาการทํางานของตับสูงขึ้น น้ําตาลในเลือดต่ําในผูปวยเบาหวานชนิดที่ตอง พึ่งอินซูลิน9, 12-13 แพทยจํานวนมากนําวิธีการรักษาดวย EDTA ดังกลาวมาใชรักษาผูปวยโรคหลอดเลือดหัวใจและ สมองอุดตัน โดยใหเหตุผลวาเปนการบําบัดแบบแพทยทางเลือกที่นิยมและยอมรับในตางประเทศ โดย EDTA ชวยใหการไหลเวียนของเลือดในหลอดเลือดแดงที่เลี้ยงกลามเนื้อหัวใจกลับเปนปกติไดโดยไมตอง ผาตัด และใชเปนการรักษาเชิงปองกัน ซึ่ง American College of Advancement in Medicine ใหการสงเสริม และสนับสนุน Chelation therapy14 ปจจุบันมีการนํา Chelation therapy มาใชเปนทางเลือกของการรักษา หรือใชเพื่อประกอบการรักษา ภาวะ atherosclerosis และลดอาการของโรคหลอดเลือดหัวใจ (coronary artery disease) ซึ่งมีการแนะนําวา ปลอดภัย และเปนวิธีที่ไมตองผาตัดเพื่อใหมีเลือดไหลเวียนในหลอดเลือดแดงที่แข็ง ดังนั้นจึงสามารถ ปองกันอาการตางๆ ของโรคหัวใจ อาทิ อาการปวดเคนหัวใจ (angina pectoris) อาการปวดขาเปนระยะ เพราะขาดเลือด (claudication) กลามเนื้อหัวใจตาย (myocardial infarction) และโรคหลอดเลือดสมอง (stroke) เปนตน ดวยเหตุนี้ องคกรที่เปนการแพทยทางเลือกหลายแหงจึงไดมีการนํา Chelation therapy มา ใชอยางกวางขวางกับผูที่มีหลอดเลือดแดงแข็ง และโรคหัวใจ อยางไรก็ตาม การใช Chelation therapy ใน ปจจุบันเพ่ือรักษา atherosclerosis และโรคหัวใจยังมีขอขัดแยงในดานประสิทธิผลของ EDTA ซึ่งขอมูล สนับสนุนยังมีไมเพียงพอ องคการอาหารและยาของสหรัฐอเมริกา (FDA) พิจารณาวา disodium EDTA ไมไดบรรจุอยูในรายการยา และไมมีหลักฐานเพียงพอที่แสดงถึงความปลอดภัยและประสิทธิผลที่เกิดขึ้นใน การนํามาใชรักษาโรคหลอดเลือดหัวใจโคโรนารีย14 ดังนั้น การทบทวนวรรณกรรมในดานประสิทธิผล และภาวะแทรกซอนของ Chelation therapy ใน การรักษาโรคหลอดเลือดหัวใจ โดยใชขอมูลจากงานวิจัยที่นาเชื่อถือเทาที่มีอยู ณ ปจจุบัน จะทําใหทราบ ขอมูลเกี่ยวกับ Chelation therapy ในการรักษาโรคหลอดเลือดหัวใจไดชัดเจนยิ่งขึ้น

วัตถุประสงค

ทบทวนองคความรเกู ยวกี่ ับประสิทธิผล และผลขางเคียงของ Chelation therapy ในการรักษาโรค หลอดเลือดหัวใจ

วิธีดําเนินการ 1. กําหนดขอบเขตของการศกษาึ 2. กําหนดกลยุทธการสืบคน 3

3. คัดเลือกเอกสารที่เกยวขี่ อง 4. วิเคราะหเอกสารทผี่ านการคดเลั ือก

เกณฑการพิจารณาคดเลั ือกเอกสารวิจัยสําหรับการทบทวน

เกณฑการค ัดเขา (inclusion criteria) • Chelation therapy ในผูปวยโรคหลอดเล ือดหัวใจ • เอกสารวิชาการประเภทการทบทวนอยางเปนระบบ (systematic review) การวิจยคลั นิ ิกชนิด randomized controlled trial (RCT) การวจิ ยคลั ินกชนิ ดิ non-randomized controlled trial และการ ทบทวน (review) เกณฑการค ัดออก (exclusion criteria) • เอกสารประเภทรายงานผูปวย • การศกษาในสึ ตวั ทดลอง

กลุมประชากรศึกษา

ผูปวยโรคหลอดเลือดหวใจทั ี่ไดรับการรกษาดั วย Chelation therapy

การวัดผลลัพธ 

ประสิทธิผลและผลแทรกซอนของ Chelation therapy ในการรกษาโรคหลอดเลั ือดหัวใจ

วิธีการสืบคนขอมูล

กําหนดกลวธิ การสี ืบคน (search strategy) ดังตารางที่ 1-2

การสืบคนขอมูลทางอิเล็กทรอนิกส

1. สืบคนจากฐานขอมูล MEDLINE โดยใช PubMed ซึ่งจํากัดการสืบคนเพื่อใหไดเอกสารการศึกษา ในมนุษยเปนภาษาอังกฤษที่เปนการวิเคราะหอภิมาน (meta-analysis) หรือ การทบทวนอยางเปนระบบ (systematic review) หรือการวิจัยทางคลินิก (clinical trial) หรือแนวทางปฏิบัติ (practice guideline) หรือ การวิจัยแบบสุมที่มีกลุมควบคุม (randomized controlled trial) หรือการทบทวน (review) โดยการกําหนด กอนการสืบคนใหจํากัดดวย Humans, Clinical Trial, Meta-Analysis, Practice Guideline, Randomized Controlled Trial, Review, English 2. สืบคนจากฐานขอมูล The Cochrane Library

ตารางที่ 1 คําสืบคนและผลที่ไดจาก PubMed ลําดับ คําที่ใชสืบคน ผลการสืบคน 1 chelation[ti] 1567 2 chelat*[ti] 7632 3 chelat*[tw] 46255 4 EDTA[tw] 26454 5 (((EDTA[tw]) OR (#3)) OR (#2)) OR (#1) 68381 6 coronary artery disease[tw] 58267 7 coronary heart disease[tw] 31563 8 CAD[tw] 14831 9 CHD[tw] 11786 10 cardiovascular[tw] 256515 11 ischemic heart disease[tw] 16402 12 (((((ischemic heart disease[tw]) OR (#10)) OR (#9)) OR (#8)) OR (#7)) OR (#6) 345057 ((((((ischemic heart disease[tw]) OR (#10)) OR (#9)) OR (#8)) OR (#7)) OR 13 615 (#6)) AND (#5)

ตารางที่ 2 คําสืบคนและผลที่ไดจาก The Cochrane Library ลําดับ คําที่ใชสืบคน ผลการสืบคน 1 MeSH descriptor Chelating Agents explode all trees 1555 2 MeSH descriptor Cardiovascular Diseases explode all trees 56984 3 (chelation):ti,ab,kw or (EDTA):ti,ab,kw 732 4 (coronary artery disease):ti,ab,kw or (coronary heart disease):ti,ab,kw or 11160 (ischemic heart disease):ti,ab,kw or (CAD):ti,ab,kw or (CHD):ti,ab,kw 5 (#1 OR #3) 2036 6 (#2 OR #4) 60517 7 (#5 AND #6) 162

ประโยชนที่คาดวาจะไดรับ 5

1. ขอมูลประสิทธิผลของการรกษาโรคหลอดเลั ือดหัวใจดวยว ิธี Chelation โดยอิงหลักฐานเชิง ประจักษ  2. ขอมูลสําหรับการศกษาวึ ิจยตั อไป

ขอจํากัดการศึกษา

การศกษาอยึ ูบนพื้นฐานการสืบคนขอมูลเฉพาะการศึกษาวิจยทั ี่เปนภาษาอังกฤษ และปรากฏอยูใน ฐานขอมูลที่สามารถเขาถึงได

ผลการดําเนินการ 1. ผลการสืบคนขอมลู จากการสืบคนขอมูลตาม Search strategy ไดเอกสารทั้งสิ้น 777 เรื่อง โดยไดจาก PubMed 615 เรื่อง และ The Cochrane library 162 เรื่อง เมื่อพิจารณาตามเกณฑการคัดเขาและคัดออก ไดเอกสารที่ตรงตาม ขอกําหนดจํานวน 7 เรื่องเปนการทบทวนอยางเปนระบบ (systematic review) จํานวน 3 เรื่อง randomized controlled trial จํานวน 2 เรื่อง การทบทวนเชิงบรรยาย (narrative vreview) จํานวน 2 เรื่อง ทั้งนี้ ไมพบการ วิเคราะหอภิมาน (meta-analysis) ดังแผนภูมิที่ 1 และตารางที่ 3

Search strategy results: 777 PubMed: 615 CENTRAL: 162

Studies excluded: 770

Full text analysis: 7

Studies included: 7

แผนภูมิที่ 1 ผลการสืบคนขอม ูล

ตารางที่ 3 ประเภทและจานวนเอกสารวํ ิชาการทผี่ านการคัดเลือกเพอทบทวนื่ ลําดับ ประเภทเอกสารวชาการิ ผลการสืบคน 1 Systematic review 3 2 Randomized controlled trial 2 3 Narrative review 2

2. ผลการศกษาึ 2.1 การทบทวนวรรณกรรมเผยแพรในป 1997 ของ Lewin15 พบวามากกวา 3 ทศวรรษมาแลวที่มี การนําเอา EDTA มาทดลองใชรักษาภาวะหัวใจเสียจังหวะ (arrhythmias) และภาวะพิษจาก digitalis ในป ค.ศ.1964 เปนที่ยอมรับในวงกวางวา EDTA มีประโยชนในการแกไขภาวะพิษจาก digitalis และกดการบีบ ตัวของหัวใจหองลางผิดปรกติแบบ ventricular premature contraction ในกรณีฉุกเฉิน อยางไรก็ตาม ขอมูล สนับสนุนการนํามาใชทางคลินิกยังมีจํานวนนอย ในการใหขอมูล Chelation therapy กับผูปวยนั้น แพทย ควรทราบขอมูลเกี่ยวกับสารที่เปน chelators ตางๆ อาทิ ประวัติความเปนมา การออกฤทธิ์ รวมถึงขอแนะนํา และไมแนะนําในการนําไปใชทางการแพทย นอกจากนี้ ผลการศึกษารายงานวา EDTA เปนพิษตอไต บาง กรณีเกิดการความเสียหายที่หลอดไตฝอยสวนตน (proximal tubule) รวมถึง EDTA มีความสัมพันธกับภาวะ น้ําตาลในเลือดต่ํา ความรูสึกไมสบาย (malaise) อาการทางระบบทางเกิดอาหาร T-wave inversion ผิวหนัง อักเสบ และอาการของภาวะแคลเซียมในเลือดต่ํา เชน ชัก เปนตน 2.2 การทบทวนอยางเปนระบบของ Ernst16 เพื่อใหไดขอมูลทางคลินิกที่สนับสนุน และคัดคาน ประสิทธิผลและสัมฤทธิผลของ Cheletion therapy สําหรับโรคหลอดเลือดหัวใจ โคโรนารีย ซึ่งเผยแพรในป 2000 ผูวิจัยสืบคนขอมูลโดยใชคอมพิวเตอรจํานวน 4 เครื่องที่เปนอิสระในการสืบคนขอมูลจาก MEDLINE, EMBASE, Cochrane Library และ CISCOM (a database specialized in complementary/ alternative medicine) ตั้งแตป  1998 เปนตนมา จากการสืบคนพบวา 22 การศึกษาที่สวนใหญเปน case reports และ case series ซึ่งไมมีกลุมควบคุม การรายงานผลสวนใหญเปนการรายงานอาการที่เปนนามธรรม ที่ดีขึ้นของคนไขสวนใหญ มีการศึกษา 2 รายที่รายงานผลอยางเปนรูปธรรมจากภาพถายหลอดเลือด ซึ่งไม พบหลักฐานวา Cheletion therapy มีประโยชน มี 2 การศึกษาท่ีเปน clinical controlled trials ที่ไมมีหลักฐาน เพียงพอในการสนับสนุนประสิทธิผลของ Chelation therapy เมื่อเทียบกับยาหลอก (placebo) อยางไรก็ตาม ตองคํานึงถึงความเสี่ยงที่เกิดขึ้น ไดแก โรคไตเรื้อรัง (renal failure) หัวใจเตนผิดจังหวะ (arrhythmias) การ ชักเกร็ง (tetany) แคลเซียมในเลือดต่ํา (hypocalcemia) น้ําตาลในเลือดต่ํา (hypoglycemia) ความดันโลหิต ต่ํา (hypotension) ไขกระดูกทําหนาที่ลดลง (bone marrow depression) ระยะเวลาเลือดออกนานขึ้น (prolonged bleeding time) ชัก (convulsions) หยุดหายใจ (respiratory arrest) โรคระบบภูมิคุมกัน (autoimmune diseases) 7

2.3 Knudtson17 และคณะ (ป 2002) ไดศึกษาการใช Chelation therapy ดวย EDTA เพื่อดูการทนตอ ภาวะขาดเลือด และวัดคุณภาพชีวิตของผูปวยโรคหัวใจขาดเลือดชนิดคงที่ (stable ischemic heart disease) โดยผูปวยมีอายุตั้งแต 21 ปขึ้นไปและเปนโรคหลอดเลือดหัวใจ ซึ่งไดรับการวินิจฉัยดวยการถายภาพทาง รังสีของหลอดเลือดหัวใจ หรือมีหลักฐานวาเปนกลามเนื้อหัวใจตาย หรือ stable angina ที่ไดรับการรักษา ดวยยา ผูวิจัยใช blocks of 10 ในการสุมแยกผูปวยเปน 2 กลุม กลุมศึกษาไดรับ 5% dextrose ในน้ํารวมกับ disodium EDTA ในขนาด 40 มิลลิกรัม/กิโลกรัม โดยมีขนาดยาสูงที่สุดไมเกิน 3 กรัม ทั้งนี้ ในสารละลายยัง ประกอบไปดวยแมกนีเซียมซัลเฟต 750 มิลลิกรัม กรดแอสคอบิค 5 กรัม และโซเดียม ไบคารบอเนต 5 กรัมในสารละลาย 5% dextrose รวมถึงใส lidocaine 80 มิลลิกรัม เพื่อลดความปวดบริเวณที่ไดรับ สารละลาย ในกลุมควบคุมจะไดรับสารละลายในลักษณะเดียวกัน แตใชน้ําเกลือ 0.9% จํานวน 20 มิลลิลิตร แทน EDTA ลักษณะของสารละลายจะไมสามารถแยกไดดวยสี หรือฉลาก ผูปวยทั้งหมดไดรับการรักษา สัปดาหละ 2 ครั้งเปนเวลา 15 สัปดาห และเพิ่มเดือนละครั้งเปนเวลา 3 เดือน ดังนั้นผูปวยจะไดรับการรักษา ทั้งสิ้น 33 ครั้ง อยางไรก็ตามผูปวยทั้งสองกลุมยังไดรับวิตามินรวมชนิดรับประทานครั้งละ 2 เม็ดวันละ 3 ครั้ง ยกเวนวันที่ไดรับการรักษา ผูวิจัยประเมินผลการศึกษาเมื่อไดรับการรักษาเปนระยะเวลา 15 และ 27 สัปดาห สิ้นสุดการศึกษาเมื่อคลื่นไฟฟาหัวใจมีการเปลี่ยนแปลง โดยพิจารณาจาก ST-segment depression อยางนอยที่สุด 1 มิลลิเมตรที่ 27 สัปดาห ผลการศึกษาพบวาผูปวยที่ผานเกณฑการทดสอบดวย treadmill จํานวน 84 คน ผานกระบวนการสุมแลวไดกลุมที่ไดรับ Chelation therapy จํานวน 41 ราย ไดรับ placebo therapy จํานวน 43 ราย คุณภาพชีวิตและคาตางๆ ที่เกี่ยวกับการออกกําลังกายของผูปวยทั้งสองกลุมไม ตางกันอยางมีนัยสําคัญทางสถิติ กลุม Chelation therapy จํานวน 1 รายมีคา creatinine ในเลือดสูงขึ้น แตไม พบความผิดปรกติของระดับเกลือแรในรางกาย 2.4 Anderson18 และคณะไดทําการศึกษาวิจัยแบบ randomized controlled trial ซึ่งไดรับการ เผยแพรในป 2003 การศึกษามีวัตถุประสงคในการประเมินผลของ Chelation therapy ที่ใช EDTA เปรียบเทียบกับยาหลอกเพื่อดูการตอบสนองของ endothelium-dependent vasomotor ในผูปวยโรคหลอด เลือดหัวใจโคโรนารีย โดยทําการคัดเลือกผูปวยจํานวน 53 รายเขารวมโครงการ PATCH (Program to Assess Alternative Treatment Strategies to Achieve Cardiac Health) ผูปวยมีอายุตั้งแต 21 ปขึ้นไปที่ไดรับ การตรวจวาเปนโรคหัวใจโคโรนารียดวยวิธีถายภาพทางรังสีของหลอดเลือดหัวใจ หรือมีหลักฐานแสดงวา เปนกลามเนื้อหัวใจตาย (myocardial infarction) หรือมีอาการเจ็บเคนหนาอกแบบคงที่ (stable angina) ที่ ไดรับการรักษาดวยยา ผูปวยไดรับการสุมดวย block of 10 โดยผูปวยจะไมทราบวาอยูในกลุมทดลอง หรือ กลุมควบคุม ผูปวย 6 คนถูกคัดออกจากการศึกษา กลุมทดลองจํานวน 24 คนจะไดรับ 5% dextrose in water ซึ่งมี EDTA เกลือแรและวิตามินรวมอยูดวย ในขณะที่กลุมควบคุมจํานวน 23 คนไดรับ 0.9% โซเดียมคลอ ไรด 20 มิลลิลิตรแทน EDTA ในสารละลาย ผูวิจัยวัดผลการศกษาจากการขยายของหลอดเลึ ือดแดง brachial จากการใชเครื่องอัลตราซาวดที่มีความคมชัดสูงวัดเสนผานศูนยกลางของ brachial artery โดยผูปวยจะไดรับ 8

การวัด 3 ระยะ ไดแก baseline หลังจากไดรับ Chelation therapy ครั้งแรก และครั้งสุดทายหลังจากไดรับการ รักษาดวยวิธี Chelation ครบ 33 ครั้ง แลวจึงนําผลที่ไดมาคํานวณหา flow-mediated vasodilation (FMD) ผล การศึกษาไมพบความแตกตางอยางมีนัยสําคัญของ FMD ระหวางกลุมที่ไดรับ Chelation therapy กับกลุม ควบคุม กลุมที่ได Chelation therapy จํานวน 2 คนตองออกจากการศึกษาเน่ืองจากระดับ creatinine ในเลือด สูง 1 คน และเกิดอาการเจ็บเคน unstable angina 1 คน 2.5 จาก Cochrane review ของ Dans19 และคณะที่ไดทําการทบทวนอยางเปนระบบในป 2002 และ update ในป 2005 จากฐานขอมูล MEDLINE, EMBASE และ CENTRAL (Cochrane Controlled Trials Register) โดยทําการทบทวนการศึกษาที่เปน randomized controlled trial (RCT) ของการใช Chelation therapy โดยการให EDTA เขาทางหลอดเลือดดําในการรักษาโรคหลอดเลือดแดงหัวใจแข็ง (atherosclerotic cardiovascular disease) เปรียบเทียบกับสารละลายไอโซทอนิก (isotonic solution) หรือน้ํากลั่น (distilled water) ซึ่งวัดผลการรักษาจากทุกสาเหตุที่ทําใหสียชีวิต การตายจากโรคหัวใจโคโรนารีย (coronary heart disease) และการตายจากโรคของหลอดเลือด (vascular disease) รวมถึงกรณีที่ไมเสียชีวิตแตเกิดอาการของ กลุมโรคหัวใจขาดเลือดเฉียบพลัน (acute coronary syndrome) อาทิ โรคกลามเนื้อหัวใจตาย และ อาการ เจ็บเคนหนาอกแบบไมคงที่ (unstable angina pectoris) ทั้งนี้ รวมถึงอาการของโรคหลอดเลือดสมอง เชน โรคหลอดเลือดสมอง (stroke) เปนตน ผลการทบทวนพบวาขอมูลที่สนับสนุน และไมสนับสนุน ประสิทธิผลของ Chelation therapy ในการใชรักษา atherosclerotic cardiovascular disease ยังมีไมเพียงพอ ควรทําการศึกษาวิจัยแบบ RCT เพิ่มขึ้น โดยเฉพาะในกลุมเสี่ยงโรคหัวใจโคโรนารีย และโรคหลอดเลือด สมอง รวมถึงการวัดผลลัพธควรวัดใหถึงคุณภาพชีวิตของผูปวย ผลขางเคียงจากการทํา Chelation therapy ท่ีพบ ไดแก วิงเวียนเปนลม อาการทางระบบทางเดินอาหาร (gastrointestinal symptoms) พบโปรตีนใน ปสสาวะ (proteinuria) แคลเซียมในเลือดต่ํา (hypocalcemia) รวมถึงผลขางเคียงจากการทําหัตถการ ไดแก หลอดเลือดดําอักเสบ (phlebitis) และปวดบริเวณที่ใหสารละลายเขาหลอดเลือด 2.6 Seely20 และคณะทําการทบทวนอยางเปนระบบในป 2005 เพื่อประเมิน Chelation therapy ใน การรักษาโรคหลอดเลือดหัวใจจาก 7 ฐานขอมูล ไดแก MEDLINE, EMBASE, CENTRAL (Cochrane Controlled Trials Register), AMED (Alternative Medicine), Alt HealthWatch, Pre-CINAHL, CINAHL และ the Nursing and Allied Health Collection รวมถึงการสืบคนขอมูลที่ไมไดเผยแพร และการศึกษาที่อยู ระหวางการดําเนินการโดยผานทางเว็บไซต clinicaltrials.gov และ National Research Register ของประเทศ อังกฤษ การศึกษาทั้งหมดเปนชนิด randomized controlled trial ที่ใช EDTA ในผูที่มีความเสี่ยงตอโรคหลอด เลือดหัวใจ การศึกษาที่เปน non-randomized trials และการวิจัยดานเภสัชจลศาสตร (pharmacokinetic studies) จะถูกคัดออก การศึกษาที่ผานการคัดเลือกเขามาจํานวน 7 เรื่องโดยเปนการศึกษาที่สามารถนํามา ทบทวนอยางเปนระบบได 5 เรื่อง (2 เรื่องเปนการนําขอมูลจากการศึกษาใหญมาแยกวิเคราะห) การ ศึกษาวิจัยที่คัดเขามาในการทบทวนอยางเปนระบบดําเนินการในประเทศสหรัฐอเมริกา แคนาดา เดนมารก 9

นิวซีแลนด และเยอรมัน ผลการศึกษาพบวามี 2 การศึกษา (n = 19) ที่แสดงวาการใช EDTA เปน Chelation agent ไดประโยชนในโรคหลอดเลือดหัวใจ ขณะที่อีก 3 การศึกษา (n = 269) ใหผลตรงขาม ผลขางเคียงที่ เชื่อวาเกิดจาก EDTA คือ พิษตอไต (kidney toxicity) 1 ราย วิงเวียนเปนลม 8 ราย (อยูในกลุมควบคุม 2 ราย) แคลเซียมในเลือดต่ํา 12 ราย (อยูในกลุมควบคุม 1 ราย) 2.7 Chappell21 ไดทบทวนวรรณกรรมเผยแพร (ป 2007) ในประเด็นการใช EDTA ที่ใชเปน Chelation therapy แทนที่การใช clopidogrel รวมกับ aspirin ในการรักษาผูปวยที่มีความเสี่ยงจากการใสสาย สวนหลอดเลือดชนิดเคลือบยา (drug-eluting stents) ภายหลังการทําหัตถการ PTCA (percutaneous transluminal coronary angioplasty) เพื่อปองกันการจับตัวของ thrombin ซึ่งจะทําใหหลอดเลือดหัวใจ กลับมาตีบไดอีก ผลการทบทวนพบหลักฐานการทํา Chelation therapy ดวย EDTA อาจปองกันการจับเปน ลิ่มเลือดซึ่งเปนผลมาจากการใส stent ได อยางไรก็ตามขอมูลในสวนนี้ยังมีจํากัด บางการศึกษาเปนลักษณะ retrospective การศึกษาที่เปน meta-analysis ไมไดใชขอมูลที่เปน RCT ทําใหความนาเชื่อถือของการศึกษา ลดลง ผูวิจยระบั ุวาไมพบผลที่ไมพึงประสงคจากการให EDTA สําหรับขอมูลการศึกษามีที่สนับสนุนวาการให EDTA ทางหลอดเลือดดําแลวเกิด ประโยชนในการรักษาโรคหลอดเลือดหัวใจ ในขณะที่อีกหลายการศึกษาใหผลในทางตรงขาม ผูวิจัยใหความเห็นวาการศึกษาวิจัยตางๆ เหลานี้วามีอํานาจการทดสอบต่ํา หลายการศึกษาไมมี การควบคุม รวมถึงการสุมกลุมตัวอยางไมมีคุณภาพ และบางการศึกษามีการเลือกรายงานผล ซึ่งไมไดตามเกณฑกําหนดของโครงการวิจัยที่ไดรับทุนสนับสนุนจาก National Institutes of Health (NIH) เพื่อประเมิน Chelation therapy (Trial to Access Chelation Therapy, TACT)

สรุปผล อภิปรายและขอเสนอแนะ

การศึกษาตางๆ มีทั้งสนับสนุนและคัดคานประสิทธิผลของ Chelation therary สําหรับโรคหลอด เลือดหัวใจ ประเด็นที่นาสนใจคือเอกสารวิชาการที่ระบุวา Chelation therapy ไมเกิดประโยชนจะเปน ลักษณะการศึกษาที่เปนการทบทวนอยางเปนระบบ (systematic review) จากเอกสารที่เปน RCT และ เอกสารที่เปนการวิจัยทางคลินิกที่มีการสุมเลือกกลุมศึกษาและกลุมเปรียบเทียบ (randomized controlled trial, RCT)15-20 ในขณะที่มีรายงานถึงผลอันไมพึงประสงคของ Chelation therapy จาก EDTA ที่พบใน การศึกษาเหลานี้ ไดแก พิษตอไต ชัก แคลเซียมในเลือดต่ํา น้ําตาลในเลือดต่ํา เปนตน ในขณะที่เอกสารที่ สนับสนุนวา Chelation therary เกิดประโยชนนั้น เปนการทบทวนเชิงบรรยาย (narrative review) และ ลักษณะการรายงานผูปวยเปนสวนใหญ21-23 และเปนที่นาสังเกตวาผูที่รายงานไวในการศึกษาวา EDTA ไดผลในการปองกันโรคหัวใจเปนผูรวมวิจัยอยูใน TACT ดวย21 ซึ่งตองพิจารณาประเด็น conflict of interest ของผูวิจัยรวมดวยไวดวย นอกจากนี้เอกสารขอมูลการศึกษาวิจัยเกี่ยวกับประสิทธิผลของ Chelation therapy 10

มีการวัดผลลัพธของการปองกัน/ รักษาโรคหัวใจที่แตกตางกัน ซึ่งไมไดเปนการวัดผลของ EDTA ที่มีผลตอ หัวใจโดยตรง การเปรียบเทียบผลระหวางการศึกษาจึงทําไดยาก และการวัดผลลัพธดวยวิธีตางๆ อาจไม สะทอนความจริงในสิ่งที่ตองการวัดได การศึกษาวิจัยเกี่ยวกับ Chelation therapy ในลักษณะ RCT ดวยระเบียบวิธีวิจัยที่ดี มีผูเขารวม โครงการจํานวนเพียงพอ เพื่อใหไดขอมูลที่ถูกตองถึงประสิทธิผลและผลขางเคียงของ Chelation therapy ไดแก การศึกษา Trial to Access Chelation Therapy (TACT)24-25 ซึ่งไดรับทุนสนับสนุนการวิจัยจาก National Institutes of Health (NIH) ดําเนินการวิจัยในหนวยงานประมาณ100 แหงในประเทศสหรัฐอเมริกา โดยเริ่มการศึกษาในป ค.ศ. 2003 และคาดการณวาจะแลวเสร็จสมบูรณในป ค.ศ. 2012 การวิจัยเปนลักษณะ

RCT โดยใหผูปวยจะไดรับ Na2EDTA (disodium ethylenediamine tetraacetic acid) เขาทางหลอดเลือดดํา เพิ่มจากการรักษามาตรฐาน เพื่อศึกษาวาผูปวยที่มีภาวะกลามเนื้อหัวใจตายมากอนจะเกิดอาการของ โรคหัวใจและเสียชีวิตในกลุมใดมากกวากันระหวางกลุมที่ไดรับ Chelation therapy รวมกับการรักษาตาม มาตรฐานเปรียบเทียบกับการรักษาตามที่กําหนดในมาตรฐานการรักษาเพียงอยางเดียว การศึกษานี้ยังไม แลวเสร็จสมบูรณ แตมีรายงานเบื้องตนวา EDTA อาจมีประสิทธิผลในการปองกันภาวะผิดปรกติของหัวใจ ในผูที่เปนโรคหลอดเลือด21 อยางไรก็ตาม Atwood26 และคณะไดรายงานไวในป ค.ศ.2008 วา American College Advanced of Medicine (ACAM) ใชความสัมพันธดานการเมืองกดดันให NIH ใหทุนในสนับสนุน การศึกษาวิจัย TACT ในโครงรางวิจัย (protocol) ใหเหตุผลและความจําเปนในการศึกษาบนพื้นฐานการ แสดงรายงานการศึกษาผูปวยผิดพลาด โดยมองขามความเสี่ยงที่มีหลักฐานชัดเจน ผูวิจัยรวมในโครงการ ประมาณ 100 คนไมมีความเหมาะสม การศึกษาขาดการแจงถึงความเสี่ยงขั้นต่ําที่สําคัญ แบบแสดงความ จํานงการเขารวมโครงการสะทอนใหเห็นขอบกพรองและความลมเหลวในการเปดเผยการมีสวนไดสวนเสีย จากการวิจัย รวมถึงผลลัพธของการศึกษาไมมีความนาเชื่อถือ และสวนใหญยังมีขอนากังขา ในประเด็นตรง ขาม Clay27 ใหความเห็นวา TACT สมควรดําเนินการตอไป หากจะหยุดการศึกษาวิจัยควรเปนเหตุผลที่วา Chelation therapy ไมสามารถชวยชีวิตผูปวยไวไดจริง อยางไรก็ตาม หากการวิจัยเปนไปตามหลักการ ปฏิบัติการวิจัยทางคลินิกที่ดี (good clinical practice, GCP) 28 เปนมาตรฐานสากลดานจริยธรรมและดาน วิชาการ โดยมีการตรวจสอบอยางเปนระบบ ทั้งการกํากับ (monitor) การตรวจสอบ (audit) และการตรวจ ตรา (inspection) ซึ่งในแตละขั้นตอนอาจแสดงใหเห็นถึงคุณภาพของการศึกษาวิจัยวาถูกตองและเหมาะสม ตาม GCP ผลการวิจัยทางคลินิกก็นาจะเชื่อถือได การวิจัยที่ดีตองสามารถตรวจสอบได คําแนะนําสําหรับการศึกษาสวนใหญระบุวา ยังไมมีเอกสารทางวิทยาศาสตรที่ชัดเจนในการแนะนํา ใหผูปวยเลือก Chelation therapy ในการรักษาโรคหัวใจ ดังนั้น แพทยควรใหขอมูลที่ถูกตองนาเชื่อถือเทาที่ มีอยูในปจจุบันโดยปราศจากผลประโยชนแอบแฝง อาทิ การไดคาตอบแทนในการสงผูปวยไปรักษาดวยวิธี ดังกลาว หรือแพทยเปนเจาของ/ หุนสวนของ Chelation therapy เปนตน ผูปวยควรไดรับทราบถึง 11

ประสิทธิผลของ Chelation therapy คาใชจายที่ผูปวยตองเสียไป รวมถึงภาวะแทรกซอนที่อาจเกิดขึ้น เนื่องจากการทํา Chelation therapy จัดไดวาเปนวิธีการที่อาจกอใหเกิดอันตรายตอผูปวยได

เอกสารอางอิง

1. Wikipedia. Chelation therapy. URL: http://en.wikipedia.org/wiki/Chelation_therapy. (accessed 21/06/2010). 2. American Cancer Society. Chelation therapy. Available from: URL:http://www.cancer.org/docroot/eto/content/eto_5_3x_chelation_therapy.asp. (accessed 21/06/2010). 3. Tabangcura D, Daubert GP. British anti-Lewisite. Available from: URL:http://www.chm.bris.ac.uk/motm/bal/index.html. (accessed 21/06/2010). 4. Wisegeek. What is chelation therapy? Available from: URL:http://www.wisegeek.com/what-is- chelation-therapy.htm. (accessed 21/06/2010). 5. Green S. Chelation therapy: unproven claims and unsound theories. Available from: URL:http://www.quackwatch.org/01QuackeryRelatedTopics/chelation.html. (accessed 21/06/2010). 6. Healthline. Chelation Therapy. Available from: URL:http://www.healthline.com/natstandardcontent/alt-edta-therapy?print=true. (accessed 21/06/2010). 7. Wellness.com. Chelation (EDTA) therapy. Available from: URL:http://www.wellness.com/reference/healthwellness/chelation-edta-therapy/general-information. (accessed 21/06/2010). 8. American Heart Association. Chelation therapy: AHA recommendation. Available from: URL:http://www.americanheart.org/presenter.jhtml?identifier=4493 (accessed June 14, 2010). 9. Anand A, Evans MF. Dose chelation therapy work for ischemic heart disease? Can Fam Physician. 2003 Mar;49:307-9. 10. Assistant Secretary of Legislation (ALS). Department of Health and Human Services. Statement on chelation therapy. Available from: URL:http://www.hhs.gov/asl/testify/t990310a.html. (accessed 21/06/2010). 11. Lamas GA, Hussein SJ. EDTA chelation therapy meets evidence-based medicine. Complement Ther Clin Pract 2006;12:213-5. 12

12. Center of Disease Control and Prevention. Deaths associated with hypocalcemia from chelation therapy --- Texas, Pennsylvania, and Oregon, 2003—2005. Available from: URL:http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5508a3.htm. (accessed 21/06/2010). 13. The Medical News. Chelation therapy side effects. Available from: URL:http://www.news- medical.net/health/Chelation-Therapy-Side-Effects.aspx. (accessed 21/06/2010). 14. Chappell LT. Applications of EDTA chelation therapy. Altern Med Rev 1997;2(6):426-32. 15. Lewin MR. Chelation therapy for cardiovascular disease. Tex Heart Inst J 1997;24(2):81-9. 16. Ernst E. Chelation therapy for coronary heart disease: An overview of all clinical investigations. Am Heart J 2000;140:139-41. 17. Knudtson ML, Wyse DG, Galbraith PD, Brant R, Hildebrand K, Paterson D, et al. Chelation therapy for ischemic heart disease: a randomized controlled trial. JAMA 2002;287(4):481-6. 18. Anderson TJ, Hubacek J, Wyse G, Knudtson ML. Effect of chelation therapy on endothelial function in patients with coronary artery disease: PATCH substudy. J Am Coll Cardiol 2003;41(3):420-5. 19. Dans AL, Tan FN, Villarruz-Sulit EC. Chelation therapy for atherosclerotic cardiovascular disease. Cochrane Database of Systematic Reviews 2002, Issue 4. Art. No.: CD002785. DOI: 10.1002/14651858.CD002785. 20. Seely DMR, Wu P, Mills EJ. EDTA chelation therapy for cardiovascular disease: a systematic review. BMC Cardiovascular Disorders 2005,5:32. 21. Chappell LT. Should EDTA chelation therapy be used instead of long-term clopidogrel plus aspirin to treat patients at risk from drug-eluting stents. Altern Med Rev 2007;12(2):152-8. 22. Quan H, Galbraith PD, Norris CM, Southern DA, King K, Verhoef MJ, et al. Opinions on chelation therapy in patients undergoing coronary angiography: cross-sectional survey. Can J Cardiol. 2007;23(8):635-40. 23. Quan H, Ghali WA, Verhoef MJ, Norris CM, Galbraith PD, Knudtson ML. Use of chelation therapy after coronary angiography. Am J Med 2001;111:686-91. 24. U.S. National Institute of Health. Trial to Assess Chelation Therapy (TACT). Available from: URL:http://clinicaltrials.gov/ct2/show/NCT00044213. (accessed 21/06/2010). 25. NIH Guide. EDTA chelation therapy for coronary artery disease. Available from: URL:http://grants.nih.gov/grants/guide/rfa-files/RFA-AT-01-004.html. (accessed 21/06/2010). 26. Atwood KC, Woeckner E, Baratz RS, Sampson WI. Why the NIH trial to assess chelation therapy (TACT) should be abandoned. Medscape J Med 2008;10(5):115. 13

27. Clay B. Study of chelation therapy should not be abandoned. Journal of American Physicians and Surgeons 2009;14(2):52-7. 28. สํานักงานคณะกรรมการอาหารและยา กระทรวงสาธารณสุข. ICH Good clinical practice (ฉบับ ภาษาไทย); 2543.

ภาคผนวก

Chelation Therapy for Cardiovascular Disease Review and Commentary

Matthew R. Lewin

Guest Editorial Chelation Therapy for Cardiovascular Disease Review and Commentary

Matthew R. Lewin T wo years ago, C. Richard Conti, in his capacity as editor-in-chief of Clini- cal Cardiology, wrote a brief editorial on chelation therapy for atherosclerosis.' His conclusion ("I counsel all patients who ask me about this therapy not to seek out individuals who prescribe it") was not startling, given the almost total absence of scientific evidence in support of the treatment. What did surprise me was the need for such an editorial in 1995, after the adoption of a dozen formal position statements by medical organizations* and the federal government. Dr. Conti acted, he explained, at the request of a cardiology colleague, and after having been asked by a number of his patients about the value of chelation therapy for coronary heart disease. A surprisingly large number of people in the United States seek out unproven and potentially harmful alternative therapies. A 1993 study published in the New EnglandJournal ofMedicine7 suggested that each year Americans make more than 400 million visits to providers of unconventional therapies, a number greater than the total number of visits to all primary care physicians. The annual out-of- pocket expenditure has been estimated to be $10 billion.8 In order to answer patients' questions about chelation therapy, a physician should know a few things about chelators: their history, their method of action, and their confirmed (and unconfirmed) medical applications. Background

Chelation therapy usually involves a series of slow-drip, intravenous infusions of the synthetic amino acid EDTA (ethylenediamine tetraacetic acid).** For many years, EDTA has been approved by the Food and Drug Administration for use in the treatment of some types of heavy metal poisoning (e.g., inorganic lead); and it was used in the treatment of digitalis poisoning until antibody fragments took Mr. Lewin is in his 5th year of training in the combined its place. The term chelate was coined by analytical chemist G.T. Morgan in 1920;10 MD/PhD Program of the it comes from the Greek chela, or "claw," which refers to the claw-like chemical University of Texas Medical structure of the compound (Fig. 1). In chelation, a metallic ion is sequestered School at Houston. He received his BS degree in and firnnly bound into a ring within the chelating molecule. Alfred Werner was entomology (1991) from the awarded the 1913 Nobel Prize in chemistry for developing this concept." University of California Natural Chelators. Naturally occurring chelating agents, such as citrate, were at Berkeley. among the 1st used in medicine. According to Jones,'2 the 1st modern medical exploitation of the properties of chelation occurred in 1917, with the use of tar- Key words: Alternative trate'2-and then, in 1925, of tironl3-to reduce the toxicity of antimonial-based medicine/utilization; atherosclerosis/drug therapy; antiparasitic agents used against diseases such as schistosomiasis.'3 In a 1938 pub- chelation therapy; chelating lication, Health Hazards in the Plating Shop: Some Suggestionsfor TheirElimina- agents; edetic acid/adminis- tion,'4 F.M. Carlsen proposed the empiric use of "fruit juices" for the treatment of tration & dosage; edetic acid/ adverse effects; edetic acid/ nickel-induced eczema. In 1943, Kety and Letonoff"5 explored the use of sodium therapeutic use; history of citrate to reduce the lead burden in human beings. medicine, modern; Versene Synthetic Chelators. Synthetic chelators are abundant and serve as useful ana- Address for reprints: lytical and biochemical tools. By the time chemist Ferdinand Munz, at the Ger- Matthew R. Lewin, Department of Integrative There have been, for example, statements (advising against chelation therapy for cardiovascular dis- Biology, MSB 4. 100, ease) by the American Medical Association (1984),2 the American College of Cardiology (1984),3 the University of Texas American Heart Association (1985),4 the American Osteopathic Association (1990),5 and the Ameri- Medical School, can Academy of Family Physicians (1993).6 6431 Fannin Street, Houston, TX 77030 "Also called Versene (or calcium disodium versenate9).

Texas Heart Instituteiournal Chelation Therapy for Cardiovascular Disease 81 0 0 the potential of chelation-based therapies, including EDTA. NaOCCH2 C-C CH2CONa Two new areas in which chelating agents are of interest in the medical community today are cerebrovascular physiology and neurobiology. The invention of membrane-permeant analogs of EDTA and related compounds (e.g., BAPTA-AM and EGTA- k ECaX AM*) has opened new frontiers in these areas of research. First synthesized in 1975 by Roger Tsien31 as an analytical reagent, BAPTA-AM has since been Fig. 1 Ethylenediamine tetraacetic acid effectively binds di- and trivalent cations, forming a stable ring. Here it has bound studied by Tymianski and coworkers to determine calcium, to form edetate calcium disodium. the mechanisms by which calcium ion chelators reduce early excitotoxic and ischemic neuronal injury in vitro and in vivo.32'33 As in the past, EDTA-based chelating agents are interesting and powerftil ana- man company General Aniline, received the U.S. lytical and research tools. Applications for cell- patent for the structure, synthesis, and application permeant analogs of calcium ion chelators may in of EDTA in 1938 (Fig. 2),16 there was already a rich the future have a major impact on a wide range of tradition in analytical chemistry surrounding chelat- medical specialties, including neurosurgery, trauma ing compounds. Synthetic methods such as the ones care, cardiology, and (especially) cardiovascular developed by Manz16 in the 1930s and Bersworth'7'18 surgery. I anticipate that a significant application in the 1940s ameliorated the chief disadvantages of for membrane-permeant calcium chelators will be earlier synthetic methods, i.e., low yields and the prophylactic reduction in the activity of calcium- release of free hydrogen cyanide.19 Applications for dependent enzymes thought to be important in post- chelating compounds already considered or in use ischemic pathology, especially iatrogenic types. These by the late 1930s included the study of metal ions in hypotheses will need to be tested rigorously, in a biochemical systems such as chlorophyll and hemo- controlled scientific setting. globin. Such compounds helped settle questions related to stereochemistry in organic systems, to the Cardiovascular theory and practice of mordant dyeing, and to phar- Applications of EDTA maceutical preparations.20 Medical Applications of Chelation. A synthetic This brings us to the cardiovascular applications of compound, dimercaprol (Fig. 3), better known as chelators. At present, the clinically proven applica- British antilewisite (BAL), was introduced in 1945 by tions are few. Over 3 decades ago, EDTA was exam- Peters and coworkers21 as a remedy for military ar- ined as a tool to treat certain arrhythmias and senical blisters and systemic arsenic poisoning from digitalis poisoning. By 1964, there was widespread the war gas lewisite.21'22 It was soon found to be agreement that EDTA was useful in the management useful against other heavy metal poisonings, such as of digitalis poisoning and in the suppression of ven- that caused by mercury. The introduction of EDTA- tricular premature contractions in the emergency based chelating agents to combat heavy metal poi- setting.34-36 soning was a tremendous advance in toxicology. To In 1955, Clarke, Clarke, and Mosher introduced put the impact of this new technology in perspec- the possibility that EDTA could be used as a treat- tive, one has only to remember that the 1st edition ment for atherosclerotic heart disease.37 Twenty-two of Goodman and Gilman's The Pharmacological patients were treated with EDTA for a variety of Basis of Therapeutics, published in 1941, states the reasons; these included several patients who were following objective in the treatment of lead poison- receiving EDTA experimentally in an effort to deter- ing: "the main objective ... is to drive the lead from mine its effects on serum calcium. Particular atten- the circulation and deposit it in the bones.... 23 tion was given to the case history of 1 patient under In the 1950s and 1960s, there was an explosion of treatment for progressively worsening bilateral neph- publications on the effects of various chelating rocalcinosis. The apparent salutary effects of EDTA agents in animals and human beings. By 1951, EDTA on this patient included improved hearing and im- was being used for the treatment of inorganic lead provement of digestive symptoms. The authors pre- poisoning24'25 and was under investigation as a tool sented radiographic evidence of improvement in this to treat hypercalcemia26-28 and even radiation poison- * 1,2-bis-(2-amino-phenoxy)ethane-N,N,N'N'-tetraacetic acid ace- ing from plutonium.29 Contemporary discoveries in toxymethyl ester (BAPTA-AM) and ethylene glycol-bis(heta- the field of cancer chemotherapy spawned an im- aminoethyl ether)-N,N,N'.N'-tetraacetic acid acetoxymethyl ester mense body of investigation (over 600 papers30) on (EGTA-AM)

82 Chelation Therapy for Cardiovascular Disease Voltime 24, Ntiniber 2, 1997 Patented Sept. 20, 1938 2130,505

UNITED STATES PATENT OFFICE 239,505 IOLYAMO CAPDOZThIC ACIDS AND PROCESS OF WANING SAME

FeeUnnad MAns, Framkort-on-the-M-l- Ger- nan. amdsut to Genera A Works, Inc., Nw YoTrk N. T.. corporatim of Ieawar No Drawing. aplicton October 22. 1936. Ser We 107.020. Divided and this ap- flcadon April 2, 1927, SerIa No. 134,737. In Germany October 30, 1935

2 (CL 260-634) This applicaton is a dlvision of my copendlUg solutlon of 10% strent are mixed with 466 parts application Ser. No. 107,020, fWed October 22. of the sodium sat of monochloraceUc acld and 1936, whlch relates to the use of amino poly- 212 parts of sodum carbonate and the mixture is carboxylic acids for avoidlng and rendering heated at 90 to 95 C. for 8 to 10 hours. Then 470 * harmess the precipitates of water-inoluble metal parts of a hydrochloric acid of 20' BE. are added. salts, particularly formed owing to the hardness Whe cool an acid of the formula: of water. HOOC.U3OCH, ICO0H My present Imvention relates to PolYamino car- boxyllc ac-ids and a Drocess of making same. " Trbey are obtainable by actIng with monochoro RoOC.n.C-11 CHOCOOB t acetic acld on a poly-mine. precipitates, which is scarcely soluble in water As a Dolyamine from which the carboxylic acids and may be recrystallied from water. are derived, there may be mentioned partcularly I claim: ethylene dlamrine. 1. Polyam.ino polycarboylic acsds correspond- US In thi manner polyamlno-polycarboxyllc acids lng to the formula: is are obtained which correspond to the gent si EOoCI:K,O C8.COOH formula: \ HOOCHEC CHSCOOX N-R HIOOCH3C CHICOOH 20 -- wherein R standx for a lower alIylene radicle. 20 HOOCHI,C ,cHCOoM 2. A polycarboxyc acJd of the formula: whereia R stands for a lower al lene radlcle. The following exa&npe wi further Illustrate HOOC-HgC CHSCO,OBI how the said invention may be carried out in 25 practice, but the invenUon is not rtricted to thi aOOCHaC CHculcooH 25 example. The parts are by wreight which Is scarcely soluble ln water and may be re- zampte crystallized from water. 60 parts of ethylene diam_lne in an aquec-" PERDNAND MtZN. Fig. 2 United States patent for ethylenediamine tetraacetic acid. patient's kidney stones and implied that this was cium, and their discussion focused on the possible direct evidence of the removal of metastatic calcium. interrelationship between the calcium and the cho- Their protocol included intravenous administration lesterol found in atheromatous plaques. They con- of 5 grams of EDTA in 500 cc of normal saline or 5% cluded that EDTA was worthy of further study and glucose over a period of 4 hours. Each patient was mentioned that they were investigating the effect of given 12 to 20 infusions over a 2- or 3-week period. EDTA on angina pectoris in some patients from this The authors anticipated a mechanism by which same group and in additional patients. These results EDTA might preferentially remove metastatic cal- would be published later. Indeed, in "Treatment of Angina Pectoris with Di- H H H sodium EDTA" (1956),38 Clarke, Clarke, and Mosher I I reported that anginal symptoms, following treatment H-C-C-C-H in accordance with their EDTA protocol, improved l I in 19 of 20 patients with coronary artery and periph- SH SH OH eral occlusive disease. They cited 2 chest radio- Fig. 3 Structure of the metal complexing agent dimercaprol graphs (1 showing mitral valve calcification) as (BAL). evidence in support of their hypothesis, but they did not publish the radiographic images. At physiologic

Te-vas Heart Institidejournal Chelation Therapy for Cardiovascular Disease 83 pH, the sodium salt of EDTA is an effective chelator and Carter47 in 1988. This study enrolled 1,974 of calcium. It was this observation, and the observa- consecutive patients and reported its results in quali- tion by others that calcium and cholesterol are asso- tative terms, such as "marked" or "good" improve- ciated in the structure of atherosclerotic plaques, ment. Ninety-four percent of patients with ischemic that led Clarke, Clarke, and Mosher to hypothesize heart disease treated with EDTA chelation therapy that EDTA could dissolve disease-causing plaques in had marked or good improvement (77% and 17%, the coronary systems of human beings.38 respectively), and 97% with peripheral vascular dis- The 1st controlled clinical trial of EDTA chelation ease had marked or good improvement (91% and therapy for cardiovascular disease, conducted by 6%, respectively). An excellent, detailed review of all Kitchell and associates in 1963,39 was subtitled "A case reports and trials was published by Grier and Reappraisal," in reference to their earlier (1961) work Meyers in 1993.5l with EDTA in human subjects.40 In 1961, they had Common criticisms of the data that appear to sup- examined 10 patients with angina and reported port EDTA chelation therapy for cardiovascular dis- subjective improvement of anginal symptoms in 9 ease have included: small sample sizes, open-label patients. In the 1963 clinical study, the authors per- formats, and the qualitative rather than quantitative formed follow-up on 28 patients with angina who means by which most studies have evaluated treat- received EDTA for at least 20 treatments. Nine other ment outcomes. In addition, while there have been patients were in a double-blinded, cross-over wing relatively few fatalities directly attributable to EDTA of the study. Of these 9, 4 were treated with EDTA infusions,"5 other adverse effects have been many and 2 of the treated patients reported subjective and varied.9 The most commonly reported adverse improvement of their symptoms after 3 months. Af- effect is pain or burning at the site of infusion.41 ter 18 months, 7 (25%) of the 28 patients receiving Moreover, EDTA has been reported to have toxic EDTA were dead, 2 (7%) were "worse," 6 (22%) were effects on the kidney,9'51'52 sometimes causing proxi- "the same," and the remaining 13 (46%) were "im- mal tubule damage, and has been associated with proved." They concluded, "At present we believe hypoglycemia, malaise, gastrointestinal symptoms, that chelation as used in this study did not benefit T-wave inversion, dermatitis, and hypocalcemic patients more than other commonly used therapeu- symptoms such as tetany.9'41'51 tic methods. It is not a useful clinical tool in the treat- The 1963 clinical trial by Kitchell39 was the 1st ment of coronary heart disease at the present time." published study that was critical of the EDTA proto- To date, in the English-language medical litera- col for atherosclerosis. In fact, other than the Kitchell ture, there has been only 1 other controlled clinical study, there appears to have been little enthusiasm trial evaluating the efficacy of EDTA in cardiovascu- or interest in this experimental therapeutic modality lar disease. The trial conducted by Guldager and in the 1960s, when only 2 small case report series colleagues4' is the only placebo-controlled, double- were published (by Lamar, in 1964 and 1966).4445 By blind study with random patient allocation. One of the early 1970s, interest in EDTA chelation therapy the endpoints studied was symptomatic relief of for cardiovascular disease had pretty much faded stable, intermittent claudication. Other endpoints among most members of the medical and scientific were pain-free walking distance on the treadmill and community. maximum walking distance. One hundred fifty-three In 1973, however, the American College for Ad- patients who averaged 65 ± 9 years in age were ran- vancement in Medicine (ACAM) was founded with domized to receive 20 infusions of EDTA over a 5- the purpose, in part, of promoting chelation ther- to 9-week period, in accordance with a regimen apy.53 This group, which comprises about 750 li- similar to those used by Clarke's group and others'3839 censed physicians, sponsors certification programs in the 1950s and 1960s. No difference between the and publishes the Journal ofAdvancement in Medi- placebo and treatment groups was noted in any cat- cine. It also has an easily accessible web site on the egory investigated. Internet. The ACAM organization has published pro- Although there has been a paucity of controlled tocols for the administration of EDTA, which call for clinical trials, case-report series are abundant, and infusions of 50 mg per kilogram of body weight. This there have been 2 open-label longitudinal studies is similar to doses used in early studies.37'8 However, and 1 open-label clinical trial. I surveyed 12 consecu- ACAM has made many modifications and additions. tive case reports3 "'42-50 that specifically examined These include the use of heparin, B-complex vita- the apparent effect of EDTA on the symptoms and mins, and megadoses of vitamin C and other vita- signs of peripheral vascular disease, coronary artery min and mineral supplements. Infusions are given disease, or both. There were a total of 2,217 patients several times per week, and a complete course of treated with EDTA. Improvement was reported for therapy can include 20 to 40 trips to the doctor at a 2,060 (93%) of patients. Of these, the most notable cost ranging from $1,600 to $4,000, or $80 to $100 is the open-label clinical trial published by Olszewer dollars per treatment. These are not covered by any

84 Chelation Therapy for Cardiovascular Disease Volume 24, Number 2, 1997 insurance companies. Additionally, lifestyle- and diet- patent on the compound ran out in the 1940s.5' (In modification programs (not dissimilar to those de- fact, there are several thousand patents on the medi- veloped, published, and validated by Dean Ornish, cal and pharmaceutical uses of EDTA.) Other pro- Larry Scherwitz, and Lance Gould and coworkers ponents accuse surgeons of suppressing chelation during the last 2 decades) are encouraged to com- therapy in order to keep their wards filled with rev- plete this therapy.5455 enue-generating bypass patients. The Internet has Claims regarding the benefits of EDTA have become an efficient and cheap way for practitioners grown since the 1950s and 1960s. Dozens of books of alternative medicine to promote their therapies have been published by proponents of chelation with no regulation. therapy. These claims include, but are not limited to, In 1954, Irvine H. Page6" outlined several meth- the treatment and prevention of atherosclerosis, ods of combating atherogenesis that he deemed coronary heart disease, angina, and peripheral vas- worthy of further investigation. He described most cular disease, the reduction of blood viscosity and others as "tentative thrusts into the unknown, some free-radical formation, and the slowing of the aging capable of being built upon, most too weak to sup- process. Claims also include "increased" sexual func- port a superstructure." He followed with a warning tion,5657 improved mental and other nervous system about the growing open market for preventive function, and relief of all types of arthritis and even health products: "none are of proved value and few lupus.56 Seldom are there claims of directly antineo- have even any semblance of experimental evidence plastic activities. to support their use as a treatment or prophylactic Proposed Mechanisms ofAction. Since the intro- in atherosclerosis. Their recommendation is based duction of EDTA as a potential treatment for cardio- on the philosophy that they do no harm and might vascular disease, proponents of this therapy have do good. This is an expensive and elusive philoso- put forward several hypotheses to explain its mecha- phy, seldom contributing to the advancement of nism of action. Among the earliest was the concept sound therapeutics, and greatly contributing to logi- that if calcium were removed from the atheroscle- cal delinquency in the minds of the prescribing phy- rotic plaque, the plaque would disintegrate, thus im- sician...." His words were prophetic. proving the patency of the arteries. In The Chelation Answer (1982)58 and The Chelation Way (1990),59 Morton Walker explains that parathyroid dysfunction TABLE 1. Excerpts from a Popular Booklet on Chelation causes precipitation of calcium into the vasculature Therapy56 and other tissue. Chelation removes this calcium. Leading chelation therapy proponent Elmer Cranton A Partial List of the Benefits of Chelation Therapy proposes in his 1990 book, Bypassing Bypass,60 that Improves liver function Removes excessive iron EDTA blocks the generation of free radicals impli- deposits cated both in lipid peroxidation and in the chain of Lowers blood cholesterol Heals poor circulation ulcers events leading to atherogenesis. Because free radi- Lowers blood fats Reduces leg cramps cals have been implicated as a final common path- Reduces blood pressure Improves vision way in so many pathological processes, the expan- Improves coronary circulation Improves varicose veins, sion of claims regarding EDTA's applications has Reduces heart irritability pigmentation been dramatic. Reduces strong heart beats Relieves angina pectoris Relieves signs of senility Commentary Diseases Aided by Chelation Therapy None of the claims listed above has been published Angina pectoris Scleroderma in peer-reviewed scientific literature, but they are Coronary arteriosclerosis Hypoglycemia legion in books and flyers published by the alterna- Psoriasis Cerebral degeneration tive medicine community (Table I). Moreover, a cur- Thrombophlebitis Lupus sory search of World Wide Web sites on any of Venom bites Diabetic gangrene several servers will pull up several hundred chela- Parkinson's disease Arteriosclerosis tion therapy web sites. Of the hundred or so that I Lead toxicity visited (Table II), only 2 sites were for skeptics and [The inside front cover, however, bears the following caveat: 1 was a statement from the American Heart Associa- "This booklet is intended only for informational purposes. It tion (Fig. 4). Indeed many of the sources that I re- only advises people to question and learn for themselves that viewed have a common complaint: the existence of there are other treatments and methods available. The author and publisher make no medical claim direct or implied. The a conspiracy to suppress EDTA-based treatments. reader should consult a licensed physician for any condition Many cite the lack of financial incentive for phar- that requires his services. "561 maceutical companies to produce EDTA, since the

Texas Heart Institutejournal Chelation Therapy for Cardiovascular Disease 85 TABLE 11. This is a sample of the web sites available simply by typing "chelation therapy" when using any of the more common Internet search engines, such as Alta Vista, Yahoo, or Web Crawler. http://www.pathwaysdc.com/9-96baer.html http://www.newagemall.com/prac/WA/yelm.htmI http://www.macontel.com/data/health/thera8l 7.htm http://www.envprevhealthctratl.com/chelther.htm http://www.coos.or.us/-signal/chel2.htm http://www.acam.org/biblio.html http://www.acam.org/cict.html http://www.vitawise.com/chethe.htm http://www.abchealth.com/chelatio.htm http://www.acam.org/pospaper.html http://www.idsweb.com/ahma/josephs/therapy.htm http://www.acam.org/edtapapr.html http://www.adcomtek.com/veins/chelo.htm http://www.acam.org/schedule.html http://www.chelation.com/ http://mel.lib.mi.us/health/health-chelation.html http://www.hrt.org/wellfac3.html http://www.amhrt.org/hs96/chelate.html http://www.ssnow.com/med2life/chelation.htm http://wheel.ucdavis.edu/-btcarrol/skeptic/chelate.html http://www.eurohost.com/docwelln/chelat.html http://www.quackwatch.com/01 QuackeryRelatedTopics/ http://healthy.net/clinic/therapy/chelat/intro/ chelation.html

Chelation therapy for cardiovascular disease has substantiated either way.) These are but a few rea- been embraced by the alternative medicine commu- sons, in addition to the absence of supporting clini- nity for over 30 years, despite its lack of support by cal data, that chelation therapy with EDTA has faced clinical evidence or by logic. Grier and Meyers5' such skepticism. point out that almost all clinical studies of chelation The overwhelming majority of chelation thera- therapy are subjective and do not apply measure- pists promote their therapies as exceptionally safe, ments of endpoints uniformly. Nor are there any natural, and nontoxic. As we've already seen, this is controls. Such criticisms do not intimidate the pro- (in the case of EDTA) patently untrue. However, the ponents of chelation. In The Chelation Way,59 Walker risks of serious harm from this therapy are relatively cites 2 authorities, H. Richard Casdorph, MD, PhD, low when the drug is administered properly to a and Elmer Cranton, MD, both of whom contend that patient with healthy kidneys.62 This has enhanced its double-blind studies are unnecessary, "since each appeal for promoters, who incur little liability by its patient legitimately receiving chelation therapy serves use. as his own control." In 1993, Monaco and Green63 noted that many There are many reasons for believing that the promoters of unproven alternative therapies warn logic of chelation therapy for atherosclerotic disease patients not to tell their regular doctors because is flawed. Calcification of plaques is a late manifes- "they won't understand" and will "interfere with tation of the disease process. Moreover, the plaque your trust in our program." Moreover, these practi- is an integral part of the arterial wall and cannot be tioners' liberal use of recognized (but little-under- reached by EDTA, which is water soluble and does stood) scientific and medical terms creates an aura not pass freely through cell membranes. The target of scientific expertise that is hard to resist.63 ion for chelation therapy, Ca2l, is not the ion most Dozens of physicians have been disciplined by strongly bound by EDTA, which has a much greater state boards for using chelation therapy in unaccept- affinity for heavier metal ions. Although EDTA does able ways. In Texas, for example, an osteopath used increase the urinary excretion of calcium, there are chelation therapy in the treatment of Alzheimer's sources of calcium much more abundant than ath- disease and was disciplined in 1990 for a "profes- erosclerotic plaques-the 2 most obvious being sional failure to practice medicine in an acceptable bone and serum. manner consistent with public health and welfare."* *Then there is the newer claim that EDTA chela- In Washington, DC, a medical doctor had his license tion therapy prevents the generation of free radicals revoked for using EDTA-based therapy to treat ath- and, thereby, lipid peroxidation. This also does not erosclerosis and was told, "EDTA therapy is not an make sense, unless chelation is combined with the accepted treatment modality used by competent administration of antioxidant vitamins such as ascor- physicians.... the use of invalidated treatment mo- bic acid. Even chelated, ferric iron has 2 free elec- dalities demonstrates a willful or careless disregard trons available to help catalyze the formation of for the health, welfare and safety of patients...."* free radicals by participating in oxidation-reduction Many of these censured doctors are considered mar- chemical reactions. At best, this treatment might do tyrs in their professional circles. no harm; but at worst, it might actually generate a * Personal communication, Saul Green, PhD, Zol Consultants, Inc. greater quantity of free radicals."3 (This has not been August 1996.

86 Chelation Therapy for Cardiovascular Disease Volume 24, Number 2, 1997 *I American Heart Heart & Stroke Association Figt.w et A-Z Guide and Sfo

air.,-g-tI rff-.Wff- "TM-A.-mr-M

CHELATION THERAPY

AHA Recommendation

The American Heart Association's Task Force on New and Unestablished Therapies reviewNed the available literature on the use of chelation (E.D.T.A., ethvlenediamine tetraacetic acid) in treating arteriosclerotic heart disease. They found no scientific evidence to demonstrate any benefit from this form of therapy. The task force recoanized that there have been no adequate published trials using currentlv approved scientific methodology to support this therapy.

Furthermore. usingy this form of unpro-en treatment may deprive patients of the wxell-established benefits from the many7 other valuable methods of treating these diseases.

Fig. 4 An excerpt from the American Heart Association's recommendation in regard to chelation for the treatment of arteriosclerotic heart disease, as it appears on the World Wide Web. Copyright © 1996 American Heart Association. Used by permission.

The Appeal ofAlternative Medicine. Unorthodox therapies because they have experienced costly and practitioners are not entirely to blame. The market impersonal care by conventional practitioners.'65 for alternative medicine is a void that cries to be A New Zealand editorial66 regarding medicine on filled. In "Aequenimitas," his famous valedictory the fringes commented that what is fringe to some is address at the University of Pennsylvania (1889), central to others and (again) cited William Osler: "A William Osler urged new graduates to ". . restrain good doctor may have practice and no theory, art your indignation when you find your pet person has and no science." Certainly science has become a triturates of the 1000th potential in his pocket, or greater part of the medical equation than it was a you discover a case of Warner's Safe Cure in the century ago, but Osler's point should be considered bedroom of your favorite patient. It must needs be seriously in attempting to understand why many that offenses of this kind come; expect them and do people prefer alternatives to conventional medicine. not be vexed." That the desperately ill will seek an Some unorthodox remedies have been found to be alternative after conventional therapies have failed effective in controlled clinical trials,66 either because is clear. But contrary to the common belief of physi- of their inherent therapeutic qualities or because of cians in the United States, most seekers of alterna- their elaborate placebo effects, which can help pa- tive therapies are not terminally ill.7'64 One reason, tients alter their lifestyles in ways that are beneficial. cited by Holohan,64 for the proliferation of alterna- However, there are to date no scientific data to tive therapies among the relatively healthy is the support the claims made by chelationists in regard simple lack of professional vigilance among physi- to cardiovascular cures. Unfortunately, the sloppy cians: the absence of "clear, consistent, and active science behind the original assumptions and obser- opposition to such approaches" is "perceived as at vations-bolstered by the natural appeal of the least tacit support of unorthodoxy." Perhaps even hypothesis-has evolved step by step from good more significantly, people have sought alternative intention to the cascade of poor medicine and sci-

Texas Heart Institulejournal Chelation Therapy for Cardiovascular Disease 87 ence that envelops chelation therapy today. Al- 16. Munz F, inventor. Assignor to General Aniline Works, Inc., though some patients do ask their physicians about New York, NY, a corporation of Delaware. Polyamino car- this debunked therapy, most are not so forthcoming boxylic acids and process of making same. US patent 2,130,505. 1938 Sep 20. about their interest in chelation. It is, therefore, in- 17. Bersworth FC, inventor. Assignor to Martin Dennis Com- cumbent upon physicians to ask patients about their pany, Newark, NJ, a corporation of New Jersey. Aliphatic use of alternative therapies and to discuss these polycarboxylic amino acids and process of making them. US therapies in a constructive manner. patent 2,407,645. 1946 Sep 17. It is probable that the majority of chelation thera- 18. Bersworth FC, inventor. Verona, NewJersey. Method of pro- the car- ducing carboxylic substituted aliphatic amines and metallic pists who uncritically advocate purported salts thereof. US patent 2,461,519. 1949 Feb 15. diovascular applications of EDTA believe in their 19. Smith R, Bullock JR, Bersworth FC, Martell AE. Carboxy- practices, just as conventional practitioners believe methylation of amines. I. Preparation of ethylenediamine in theirs. However, as it is practiced today, chelation tetraacetic acid. J Org Chem 1949;14:355-61. therapy for cardiovascular disease at best defies rec- 20. Diehl H. The chelate rings. Chem Rev 1937;21:39-111. ommendation and, at worst, is a direct or indirect 21. Peters RA, Stocken LA, Thompson RHS. British anti-Lewisite (BAL). Nature 1945;156:616-9. danger to patients. How many patients have been 22. Randall RV, Seeler AO. Medical progress. BAL. N Engl J Med deprived of safer, more efficacious therapies because 1948;239:1004-9. they have sought unproven and potentially danger- 23. Goodman LS, Gilman A. The pharmacological basis of thera- ous alternative therapies? The maintenance of good peutics: a textbook of pharmacology, toxicology and thera- health, good patient care, and good scientific under- peutics for physicians and medical students. 1st ed. New York: MacMillan Co., 1941:746. standing requires perpetual vigilance. 24. Bessman SP, Ried H, Rubin M. Treatment of lead enceph- alopathy with calcium disodium versenate. Report of a case. References Med Ann DC 1952;21:312-5. 25. Rubin M, Gignac S, Bessman SP, Belknap EL. Enhancement 1. Conti CR. Chelation therapy for atherosclerosis: one man's of lead excretion in humans by disodium calcium ethylene- view [editorial]. Clin Cardiol 1995;18:545. diamine tetraacetate. Science 1953; 117:659-60. 2. American Medical Association. Proceedings of the House of 26. Spencer H, Vankinscott V, Lewin I, Laszlo D. Removal of cal- Delegates. Report of the Council on Scientific Affairs in re- cium in man by ethylenediamine tetra-acetic acid. A meta- gard to chelation therapy; 1984 December 2-5:272-4. bolic study. J Clin Invest 1952;31:1023-7. 3. American College of Cardiology: Policy statement (Dec 27. Holland JF, Danielson E, Sahagian-Edwards A. Use of ethyl- 1984). ene diamine tetra acetic acid in hypercalcemic patients. Proc 4. American Heart Association: Questions and answers about Soc Exp Biol Med 1953;84:359-64. chelation therapy (1985). 28. Bellin J, Laszlo D. Metabolism of Ca45 in man. Science 1953; 5. American Osteopathic Association: Resolution on chelation 117:331-4. therapy (issued 1985; revised and reaffirmed 1990). 29. Foreman H, Fuqua PA, Norwood WD. Experimental admin- 6. American Academy of Family Physicians: Policy statement istration of ethylenediamine-tetraacetic acid in plutonium (Jan 1993). poisoning. AMA Arch Ind Hyg Occup Med 1954;10:226-31. 7. Eisenberg DM, Kessler RC, Foster C, Norlock FE, Calkins DR, 30. Furst A. Chemistry of chelation in cancer. Springfield (IL): Delbanco TL. Unconventional medicine in the United States. Charles C. Thomas, 1963:viii,105-38. Prevalence, costs, and patterns of use. N Engl J Med 1993; 31. Tsien R. Special report: fluorescence imaging creates a win- 328:246-52,282-3. dow on the cell. Chem Eng News 1994;72(July 18):34. 8. Subcommittee on Health and Long-Term Care of the Select 32. Tymianski M, Wallace MC, Spigelman I, Uno M, Carlen PL, Committee on Aging. House of Representatives, 98th con- Tator CH, Charlton MP. Cell-permeant Ca2+ chelators reduce gress. Quackery: a $10 billion scandal. Washington, DC: early excitotoxic and ischemic neuronal injury in vitro and Government Printing Office, 1984. Committee publication in vivo. Neuron 1993;11:221-35. no. 98-435. 33. Tymianski M, Charlton MP, Carlen PL, Tator CH. Properties 9. Calcium disodium versenate. In: Physicians' desk reference. of neuroprotective cell-permeant Ca2+ chelators: effects on 51st edition. Montvale, NJ: Medical Economics, 1997:1548- [Ca2+]i and glutamate neurotoxicity in vitro. J Neurophysiol 49. 1994;72:1973-92. 10. Morgan GT, Drew HDK. Researches on residual affinity and 34. Soffer A, Chenowith M, Eichhorn GL, Rosoff B, Rubin M, co-ordination. Part II. Acetylacetones of selenium and tellu- Spencer H. Chelation therapy. Springfield (IL): Charles C. rium. J Chem Soc (Trans) 1920;117:1456-65. Thomas, 1964. 11. Kauffman GB. Alfred Werner: Founder of coordination 35. Sapeika N. Antagonism of digitalis action by ethylenedia- chemistry. Berlin: Springer-Verlag, 1966. mine tetraacetic acid. Arch Int Pharmacodyn 1954;97:373-8. 12. Jones MM. Chemistry of chelation: chelating agent antago- 36. Gubner RS, Kallman H. Treatment of digitalis toxicity by nists for toxic metals. In: Goyer RA, Cherian MG, editors. chelation of serum calcium. Am J Med Sci 1957;234:136-43. Toxicology of metals: biochemical aspects. New York: 37. Clarke NE, Clarke CN, Mosher RE. The "in vivo" dissolution Springer-Verlag, 1995:279-304. of metastatic calcium. An approach to atherosclerosis. Am J 13. Schmidt H. Antimon in der Arzeimittelsynthese. Z Angnew Med Sci 1955;229:142-9. Chem 1930;43:963. 38. Clarke NE, Clarke CN, Mosher RE. Treatment of angina pec- 14. Carlsen FM. Health hazards in the plating shop. Some sug- toris with disodium ethylene diamine tetraacetic acid. Am J gestions for their elimination. Metal Ind (Lond) 1938;52: Med Sci 1956;232:654-66. 511-2. 39. Kitchell JR, Palmon F Jr, Aytan N, Meltzer LE. The treatment 15. Kety SS, Letonoff TV. The treatment of lead poisoning by of coronary artery disease with disodium EDTA. A reap- sodium citrate. Am J Med Sci 1943;205:406-14. praisal. Am J Cardiol 1963;11:501-6.

88 Chelation Therapy for Cardiovascular Disease Volume 24, Number 2, 199 7 40. Kitchell JR, Meltzer LE, Seven MJ. Potential uses of chelation several elements in healthy subjects. Br J Clin Pharmacol methods in the treatment of cardiovascular diseases. Prog 1991;31:347-9. Cardiovasc Dis 1961;3(4):338-49. 63. Monaco GP, Green S. Recognizing deception in the promo- 41. Guldager B, Jelnes R, Jorgensen SJ, Nielsen JS, Klaerke A, tion of untested and unproven medical treatments. N Y State Morganson K, et al. EDTA treatment of intermittant claudi- J Med 1993;93(2):88-91. cation-a double-blind, placebo-controlled study. J Intern 64. Holohan TV. Referral by default. The medical community Med 1992;231:261-7. and unorthodox therapy. JAMA 1987;257:1641-2. 42. Meltzer LE, Ural ME, Kitchell JR. The treatment of coronary 65. Murray RH, Rubel AJ. Physicians and healers-unwitting artery disease with disodium EDTA. In: Seven MJ, Johnson partners in health care. N Engl J Med 1992;326:61-4. LA, editors. Metal-binding in medicine: proceedings of a 66. Cole D, St. George I. Medicine at the fringes. N Z MedJ 1993; symposium sponsored by Hahnemann Medical College and 106:130-3. Hospital: Philadelphia. Philadelphia: JB Lippincott, 1960:132- 6. 43. Clarke NE Sr, Clarke NE Jr, Mosher RE. Treatment of occlusive vascular disease with disodium EDTA. Am J Med Sci 1960;239:732-44. 44. Lamar CP. Chelation therapy of occlusive arteriosclerosis in diabetic patients. Angiology 1964;15:379-95. 45. Lamar CP. Chelation endarterectomy for occlusive atherosclerosis. J Am Geriatr Soc 1966;14:272-94. 46. Anonymous. EDTA chelation therapy for arteriosclerotic heart disease. Med Lett Drugs Ther 1981;23:51. 47. Olszewer E, Carter JP. EDTA chelation therapy in chronic degenerative disease. Med Hypotheses 1988;27:41-9. 48. Godfrey ME. EDTA chelation as a treatment of arteriosclerosis [letter]. N Z MedJ 1990;103:162-3. 49. Deucher GP. Antioxidant therapy in the aging process. In: Emerit I, Chance B, editors. Free radicals and aging. Basel: Birkhauser Verlag, 1992:428-37. 50. Casdorph HR, Farr CH. EDTA chelation therapy III: treatment of peripheral arterial occlusion, an alternative to amputation. J Holistic Med 1981;3:101-17. 51. Grier MT, Meyers DG. So much writing, so little science: a review of 37 years of literature on edetate sodium chelation therapy. Ann Pharmacother 1993;27:1504-9. 52. Klaassen CD. Heavy-metal antagonists. Edetate calcium disodium. In: Gilman AG, Rall TW, Nies AS, Taylor P, editors. Goodman and Gilman's the pharmacological basis of therapeutics. 8th ed. New York: Pergamon Press, 1990:1607-8. 53. Green S. Chelation therapy: unproven claims and unsound theories. Nutr Forum 1993;10(5):33-7. 54. Ornish D. Can lifestyle changes reverse coronary heart disease? World Rev Nutr Diet 1993;72:38-48. 55. Ornish D, Scherwitz LW, Doody RS, Kesten D, McLanahan SM, Brown SE, et al. Effects of stress management training and dietary changes in treating ischemic heart disease. JAMA 1983;249:54-9. 56. Gutting RD. Chelation therapy. A short course on description, use, functions. Topeka (KS): Tecbook Publications, 1979:1. 57. Wynn JE, Van't Riet B, Borzelleca JF. The toxicity and phar- macodynamics of EGTA: oral administration to rats and comparisons with EDTA. Toxicol Appl Pharmacol 1970;16: 807-17. 58. Walker M, Gordon G. The chelation answer: how to prevent hardening of the arteries and rejuvenate your cardiovascular system. New York: M. Evans & Co., 1982. 59. Walker M. The chelation way. The complete book of chelation therapy. Garden City Park (NY): Avery Publishing, 1990: 26-7,70-1. 60. Cranton EM, Brecher A. Bypassing bypass: the new technique of chelation therapy. 2nd ed. New York: Stein and Day, 1990. 61. Page IH. The Lewis A. Connor Memorial Lecture. Atheroscle- rosis. An introduction. Circulation 1954;10:1-27. 62. Allain P, Mauras Y, Premel-Cabic A, Islam S, HerveJP, Cledes J. Effects of an EDTA infusion on the urinary elimination of

Texas Heart InstituteJou rnaI Chelation Therapy for Cardiovascular Disease 89 Chelation therapy for coronary heart disease: An overview of all clinical investigations

E.Ernst,MD,PhD,FRCP (Edin)Exeter.United Kingdom

Chelation therapy for coronary heart disease: An overview of all clinical investigations E. Ernst, MD, PhD, FRCP (Edin) Exeter, United Kingdom

Background Chelation therapy is popular in the United States. The question of whether it does more good than harm remains controversial. Aim The aim of this systematic review was to summarize all the clinical evidence for or against the effectiveness and efficacy of chelation therapy for coronary heart disease. Methods A thorough search strategy was implemented to retrieve all clinical investigations regardless of whether they were controlled or uncontrolled. Results The most striking finding is the almost total lack of convincing evidence for efficacy. Numerous case reports and case series were found. The majority of these publications seem to indicate that chelation therapy is effective. Only 2 controlled clinical trials were located. They provide no evidence that chelation therapy is efficacious beyond a powerful placebo effect. Conclusion Given the potential of chelation therapy to cause severe adverse effects, this treatment should now be considered obsolete. (Am Heart J 2000;140:139-41.)

See related Editorial on page 4. Methods Four independent, computerized literature searches were performed, all from earliest possible available data to the end Chelation therapy can be defined as the use of repeated of 1998 (Medline, Embase, Cochrane Library, and CISCOM [a intravenous administration of ethylenediamine tetraacetic database specialized in complementary/alternative medicine]). acid (EDTA), usually in combination with vitamins, trace The aim was to identify all clinical investigations of chelation elements, and iron supplements, as a treatment for a vari- therapy for cardiovascular disease. The key words used were ety of diseases.1 chelation, EDTA, ethylenediamine tetraacetic acid, edeteate, The treatment is of proven efficacy for heavy metal edeteate sodium, edetic acid, vascular disease, and controlled poisoning.2 In “alternative” medicine it is also advocated clinical trials. Publications found were screened for further relevant articles in their bibliographies. In particular, 2 critical for the treatment of vascular disease.3 This approach is reviews5,6 and 1 meta-analysis of (mostly uncontrolled) clinical popular in the United States. Hundreds of thousands of trials7 proved to be useful sources for further references. In people currently undergo chelation therapy every year. addition, leading experts and national societies were contacted More than 1000 US physicians support chelation ther- and asked for further material. Finally, my own extensive files apy. The average cost for a course of 20 to 30 treat- were searched for relevant publications. There were no ments is around $3000.4 A recent systematic review of restrictions according to language of publication. all randomized trials1 concluded that chelation therapy Initially, it was intended that studies would be admitted if for peripheral arterial occlusive disease is not superior they related to controlled, clinical trials of intravenous chela- to placebo. This article addresses the question of tion therapy for CHD. Because of the extreme paucity of such whether chelation therapy is effective as a treatment for publications it was subsequently decided also to include all clinical investigations into this systematic review. Data were coronary heart disease (CHD). extracted in a predefined, standardized format (Tables I and II).

Results From the Department of Complementary Medicine, School of Postgraduate Medi- Twenty-two studies8-29 without control groups were cine and Health Sciences, University of Exeter. located (Table I). The majority of these studies were Submitted August 16, 1999; accepted December 17, 1999. Reprint requests: E. Ernst, MD, Department of Complementary Medicine, School of published in the 1960s and report subjective symptomatic Postgraduate Medicine and Health Sciences, University of Exeter, 25 Victoria Park improvements in most patients. The 2 case reports that Rd, Exeter EX2 4NT, United Kingdom. included objective angiographic results25,26 showed no E-mail: [email protected] evidence of benefit. Copyright © 2000 by Mosby, Inc. 17,30 0002-8703/2000/$12.00 + 0 4/1/107548 Only 2 controlled clinical trials were located. doi:10.1067/mhj.2000.107548 Kitchell et al17 conducted a placebo-controlled, double- American Heart Journal 140 Ernst July 2000

Table I. Uncontrolled studies and case reports of chelation therapy for CHD Sample Outcome Reference size measures Result

Clarke et al (1955)8 22 Symptoms Some improvements Clarke et al (1956)9 20 Symptoms 19 improved, 1 died Boyle et al (1957)10 20 Symptoms, ECG Significant improvements Clarke (1960)11 700 Symptoms 87% improved Meltzer (1960)12 10 Symptoms, ECG 9 improved, 5 improved Clarke et al (1960)13 76 Symptoms 58 improved Kitchell et al (1961)14 10 Symptoms 9 improved Boyle et al (1961)15 10 Symptoms, ECG 9 improved Meltzer et al (1961)16 81 Not stated “Therapeutically effective” Kitchell et al (1963)17 28 Symptoms, ECG 18 improved, 1 died, 13 improved Lamar (1964)18 15 Symptoms 15 improved Lamar (1966)19 3 Symptoms 1 improved, 1 died Evers (1979)20 3000 Symptoms >90% improved AAMP (1981)21 18 Symptoms 17 improved Casdorph (1981)22 18 Ejection fraction 17 improved Robinson (1982)23 248 Symptoms, ECG Significant improvements Olszewer and Carter (1988)24 844 Symptoms 821 improved McGillen (1988)25 1 Angiography No evidence of benefit Winebaugh et al (1990)26 1 Angiography No evidence of benefit Deycher (1992)27 215 Symptoms 70% improvement Hancke and Flytlie (1992)28 42 Necessity of surgery 39 were able to cancel their surgery Hancke and Flytlie (1993)29 470 Symptoms Significant improvements

AAMP, American Academy of Medical Preventics; ECG, electrocardiogram.

blind, crossover study on 9 patients with “historically has positive effects. These reports are consistent with a unquestionable” CHD.17 Patients were allocated to powerful placebo effect of chelation treatment. The receive 20 injections of placebo or EDTA over 3-month power of the placebo responses in CHD is perhaps best periods. The authors fail to mention the dosage used in highlighted by the fact that sham surgery has been shown this study. Therapeutic success or failure was assessed to yield symptomatic benefit in the vast majority of through performance on a treadmill. The authors did CHD patients.31 When angiographic verification is not analyze their results statistically but explain that 2 sought in case studies of chelation therapy,25,26 the of the 4 patients receiving EDTA in the first treatment results are negative. phase “benefitted after 6 and 12 weeks.” Only 2 patients The most impressive, albeit disappointing, finding volunteered to be treated with EDTA in the second of this systematic review is the almost total lack of con- treatment phase. None of them showed any benefit. trolled clinical trials. In view of the claims made for The second controlled trial was published as an chelation therapy and the often aggressive marketing of abstract only.30 16 male patients with angiographically this therapy,32 this void seems embarrassing. The 2 verified CHD were randomly assigned to receive EDTA reports of controlled trials are far from being fully satis- or placebo injections. Twenty treatments were given factory. One study17 was extremely small and lacked every third day and consisted of 3 to 4 g sodium EDTA sufficient detail for an objective assessment. The sec- or saline. In the course of this series, subjective improve- ond trial30 seems to be of high methodologic standard ments and an increase in physical performance were noted but has not been reported in sufficient detail to allow a in both groups with no intergroup differences (numerically fair evaluation. the increase was larger in the placebo group). Angio- The risks associated with chelation therapy are sub- graphically there was a slow progression of the coronary stantial: renal failure, arrhythmias, tetany, hypocal- sclerosis in both groups, again with no intergroup cemia, hypoglycemia, hypotension, bone marrow differences. depression, prolonged bleeding time, convulsions, respiratory arrest, and autoimmune diseases have all been described.1,3,5,26,33-35 Several deaths have been Discussion reported with little doubt about the causal role of These reports collectively provide no reliable evi- chelation therapy.36 Proponents of chelation therapy dence to suggest that chelation therapy is of benefit in might argue that, with more refined treatment regimens, CHD. Most of the uncontrolled data imply that chelation the problems of the early days have been overcome. American Heart Journal Volume 140, Number 1 Ernst 141

Yet, with the absence of reliable data to support it, this 16. Meltzer LE, Kitchell JR, Palmon F. The long-term use, side-effects, notion is unconvincing. and toxicity of disodium EDTA. Am J Med Sci 1961;245:51-7. On balance, therefore, chelation therapy cannot be 17. Kitchell JR, Palmon F, Ayton N, Meltzer LE. The treatment of coronary recommended as a useful treatment for CHD. Numerous artery disease with disodium EDTA. Am J Cardiol 1963;11:501-6. uncontrolled investigations report symptomatic improve- 18. Lamar CP. Chelation therapy of occlusive arteriosclerosis in diabetic ment, whereas the only 2 controlled trials suggest that patients. Angiology 1964;15:379-94. 19. Lamar CP. Chelation endarterectomy for occlusive atherosclerosis. J these benefits are caused by placebo effects. In view of Am Geriatr Soc 1966;14:272-94. its potential risks, chelation therapy for CHD should be 20. Evers R. Chelation of vascular atheromatous disease (experience 37 discarded in favor of therapies of proven efficacy. with 3000 patients). Laguna Hills (CA): American College for Advancement in Medicine; 1979. 21. The American Academy of Medical Preventics. Chelation therapy. References Med Lett 1981;23:51. 1. Ernst E. Chelation therapy for peripheral arterial occlusive disease. 22. Casdorph MR. EDTA chelation therapy, efficacy in arteriosclerotic Circulation 1997;96:1031-3. disease. J Holistic Med 1981;3:53-9. 2. Lin J-L, Ho H-H, Yu C-C. Chelation therapy for patients with elevated 23. Robinson DM. Chelation therapy. NZ Med J 1982;95:750. body lead burden and progressive renal insufficiency. A randomized, 24. Olszewer E, Carter JP. EDTA chelation therapy in chronic degener- controlled trial. Ann Intern Med 1999;130:7-13. ative disease. Med Hypothesis 1988;4:91-5. 3. Chappell LT, Janson M. EDTA chelation therapy in the treatment of 25. McGillen MJ, Mancini GBJ. Inefficacy of EDTA chelation therapy vascular disease. J Cardiovasc Nurs 1996;10:78-86. for coronary atherosclerosis. N Engl J Med 1988;318:1618-9. 4. Holmes NH. Chelation therapy. In: Nurse’s handbook of alternative 26. Winebaugh SR, Geraets DR. Apparent failure of edetic acid chela- and complementary therapies. Springhouse; 1999. p. 378. tion therapy for the treatment of coronary atherosclerosis. DICP 5. Grier MT, Meyers DG. So much writing, so little science: a review Ann Pharmacother 1990;24:22-5. of 37 years of literature on edetate sodium chelation therapy. Ann 27. Deycher GP. Antioxidant therapy in the ageing process. In: Emerit I, Pharmacother 1993;27:1504-9. Charles B, editors. Free radicals in aging. Basel: Birkauser Verlag; 6. Rathmann KL, Golightly LK. Chelation therapy of atherosclerosis. 1992. p. 93-117. Drug Int Clin Pharm 1984;18:1000-3. 28. Hancke C, Flytlie K. Manipulation with EDTA. Ugeskar Laeger 1992; 7. Chappell LT, Stahl JP. The correlation between EDTA chelation ther- 154:2213-5. apy and improvement in cardiovascular function: a meta-analysis. J 29. Hancke C, Flytlie K. Benefits of EDTA chelation therapy in arterioscle- Advancement Med 1993;6:139-60. rosis: a retrospective study of 470 patients. J Advancement Med 8. Clarke NE, Clarke CN, Mosher RE. The “in vivo” dissolution of 1993;6:161-71. metastatic calcium: an approach to atherosclerosis. Am J Med Sci 30. Hopf R, Gleußner M, Babej-Dölle R, et al. Wirksamkeit von chelat 1955;229:142-9. bei patienten mit koronarer herzkrankheit [abstract]. Z Kardiol 9. Clarke NE, Clarke CN, Mosher RE. Treatment of angina pectoris 1997;76(suppl 2):31. with disodium EDTA. Am J Med Sci 1956;232:654-66. 31. Dimond EG, Kittle CF, Crockett JE. Comparison of internal mammary 10. Boyle AJ, Jasper JJ, McCormick H. Studies in human and induced ligation and sham operation for angina pectoris. Am J Cardiol atherosclerosis employing EDTA. Bull Schw Akad Wiss 1960;5:483-6. 1957;13:408-25. 32. Chappell LT, Wilson J. Chelation therapy for vascular disease. Dallas: 11. Clarke NE. Arteriosclerosis, occlusive vascular disease and EDTA. American Heart Association; 1999. p. 164. Am J Cardiology 1960;2:233-6. 33. Anonymous. Chelation therapy. JAMA 1983;250:672. 12. Meltzer LE. Chelation therapy. In: Seven MJ, Johnson LA, editors. Metal- 34. Peterson GR. Adverse effects of chelation therapy. JAMA 1983; binding in medicine. Philadelphia: JB Lippincott; 1960. p. 132-48. 250:2926. 13. Clarke NE Sr, Clarke NE Jr, Mosher RE. Treatment of occlusive 35. Meyer FP. Über die omnipotenz der chelattherapie. Forschende vascular disease with disodium EDTA. Am J Med Sci 1960;239: Komplementär 1998;5:226-71. 732-44. 36. Magee R. Chelation treatment of atherosclerosis. Med J Aust 1985; 14. Kitchell JR, Meltzer LE, Seven MJ. Potential uses of chelation methods 142:514-5. in the treatment of cardiovascular diseases. Prog Cardiovasc Dis 37. Haskell WL, Luskin FM, Marvasti FF. Complementary/alternative 1961;3:338-49. therapies in general medicine: cardiovascular disease. In: Comple- 15. Boyle AJ, Clarke NE, Mosher RE, McCann DS. Chelation therapy mentary/alternative medicine, an evidence-based approach. Spencer in circulatory and sclerosing diseases. Fed Proc 1961;29:243-51. JW, Jacobs JJ, editors. St Louis (MO): Mosby; 1999. p. 94. Chelation Therapy for Ischemic Heart Disease A Randomized Controlled Trial

Merril L.Knudtson,D.George Wyse,P.Diane Galbraith,et al.

Chelation Therapy for Ischemic Heart Disease: A Randomized Controlled Trial

Merril L. Knudtson; D. George Wyse; P. Diane Galbraith; et al. Online article and related content current as of July 29, 2009. JAMA. 2002;287(4):481-486 (doi:10.1001/jama.287.4.481)

http://jama.ama-assn.org/cgi/content/full/287/4/481

Correction Contact me if this article is corrected. Citations This article has been cited 19 times. Contact me when this article is cited. Topic collections Randomized Controlled Trial Contact me when new articles are published in these topic areas. Related Articles published in January 23/30, 2002 the same issue JAMA. 2002;287(4):527.

Subscribe Email Alerts http://jama.com/subscribe http://jamaarchives.com/alerts

Permissions Reprints/E-prints [email protected] [email protected] http://pubs.ama-assn.org/misc/permissions.dtl

Downloaded from www.jama.com by guest on July 29, 2009 ORIGINAL CONTRIBUTION

Chelation Therapy for Ischemic Heart Disease A Randomized Controlled Trial

Merril L. Knudtson, MD Context Chelation therapy using EDTA is an unproven but widely used alternative D. George Wyse, MD, PhD therapy for ischemic heart disease. P. Diane Galbraith, BN Objective To determine if current EDTA protocols have a favorable impact on exercise ischemia threshold and quality of life measures in patients with stable ischemic Rollin Brant, PhD heart disease. Kathy Hildebrand, BN Design Double-blind, randomized, placebo-controlled trial conducted between Janu- Diana Paterson, BScN ary 1996 and January 2000. Deborah Richardson, RN Setting Participants were recruited from a cohort of cardiac catheterization patients and the practices of cardiologists in Calgary, Alberta. Connie Burkart, BN Participants We screened 3140 patients, performed a qualifying treadmill test in Ellen Burgess, MD 171, and enrolled 84. Entry criteria included age at least 21 years with coronary artery for the Program to Assess Alternative disease proven by angiography or a documented myocardial infarction and stable an- Treatment Strategies to Achieve gina while receiving optimal medical therapy. The required treadmill test used a gradual Cardiac Health (PATCH) ramping protocol and patients had to demonstrate at least 1-mm ST depression. Investigators Interventions Patients were randomly assigned to receive infusion with either weight- adjusted (40 mg/kg) EDTA chelation therapy (n=41) or placebo (n=43) for 3 hours SCHEMIC HEART DISEASE CONTINUES per treatment, twice weekly for 15 weeks and once per month for an additional 3 months. to be the leading cause of death and Patients in both groups took oral multivitamin therapy as well. disability among North American Main Outcome Measure Change from baseline to 27-week follow-up in time to adults. Testimonials of symptom- ischemia (1-mm ST depression). Iatic improvement frequently lead pa- Results Thirty-nine patients in each group completed the 27-week protocol. One tients with ischemic heart disease to seek chelation patient had therapy discontinued for a transient rise in serum creatinine. The alternative therapies that have not been mean (SD) baseline exercise time to ischemia was 572 (172) and 589 (176) seconds in scrutinized in clinical trials. One such the placebo and chelation groups, respectively. The corresponding mean changes in therapy is the repeated intravenous ad- time to ischemia at 27 weeks were 54 seconds (95% confidence interval [CI], 23-84 ministration of the chelating agent EDTA seconds; PϽ.001) and 63 seconds (95% CI, 29-95 seconds; PϽ.001), for a difference in combination with oral vitamins and of 9 seconds (95% CI, −36 to 53 seconds; P=.69). Exercise capacity and quality of life minerals. Two small randomized con- scores improved by similar degrees in both groups. trolled clinical trials showed no benefit Conclusion Based on exercise time to ischemia, exercise capacity, and quality of life of EDTA in patients with peripheral ar- measurements, there is no evidence to support a beneficial effect of chelation therapy in terial disease,1,2 and the few available tri- patients with ischemic heart disease, stable angina, and a positive treadmill test for ischemia. als in ischemic heart disease are unin- JAMA. 2002;287:481-486 www.jama.com formative.3-5 Although the recent review by Ernst6 concluded that chelation nadian study, 8% of patients who had lion. The actual amount is likely higher therapy for coronary heart disease undergone cardiac catheterization and because these estimates do not in- should be considered “obsolete,” many responded to a survey had used chela- clude all the cardiac patients who do patients continue to seek alternative tion therapy.9 Assuming 8% of the 1.25 therapy.7 million US residents who have under- 10 Author Affiliations: Division of Cardiology, Univer- There are no current data on the gone cardiac catheterization have tried sity of Calgary and Calgary Regional Health Author- number of patients seeking chelation chelation therapy, we project that ity, Calgary, Alberta. 8 A list of the PATCH Investigators appears at the end therapy. In 1993, Grier and Meyers es- 100000 have tried chelation therapy. of this article. timated that more than 500000 people Estimating a cost of $4000 for the usual Corresponding Author and Reprints: Merril L. Knudt- son, MD, Foothills Medical Center, 140329th St NW, in the United States are treated with series of treatment sums to an annual Calgary, Alberta, Canada T2N 2T9 (e-mail: EDTA therapy each year. In a recent Ca- expenditure of approximately $400 mil- [email protected]).

Downloaded from www.jama.com by guest on July 29, 2009 CHELATION THERAPY FOR ISCHEMIC HEART DISEASE not undergo catheterization and all utes. A 12-lead ECG was recorded ev- atinine were measured at every fifth other noncardiac patients who seek che- ery 20 seconds. Maximum oxygen visit. Hematology, electrolyte, and cho- · lation therapy. consumption (VO2max) and anaero- lesterol panels were measured at base- We undertook a randomized, double- bic thresholds were determined by con- line and at 15 and 27 weeks. The study blind, placebo-controlled clinical trial tinuous measurement of expired gases nurse supervised patients throughout of chelation therapy using the Ameri- using a gas analyzer (MedGraphics the duration of therapy, and hourly can College for Advancement in Medi- model CPX/D; MedGraphics Corpora- pulse and blood pressure measure- cine (ACAM) protocol11 to determine tion, St Paul, Minn) calibrated online. ments were obtained. the efficacy of EDTA with respect to exercise ischemia threshold, symptoms, Randomization and Treatment Study End Points and quality of life in patients with stable Patients were randomized in blocks of Exercise parameters and quality of life ischemic heart disease. 10. Investigators were blinded to treat- questionnaires were collected at 15 and ment assignment. The hospital phar- 27 weeks after randomization. The pri- METHODS macy assigned the randomized therapy mary end point was the change in time Study Subjects and prepared solutions for blinded ad- to reach at least 1 mm of ST-segment de- Patients were recruited from the Al- ministration of infusions. The 500-mL pression at the 27-week evaluation. Pa- berta Provincial Project for Outcome infusion solution of 5% dextrose in wa- tients who did not achieve ischemic Assessment in Coronary Heart Dis- ter for the active treatment containing changes at 27 weeks had the test pe- ease (APPROACH) cohort of cardiac disodium EDTA (Endrate; Abbott Labo- riod truncated at 14 minutes, and 14 catheterization patients12 and the prac- ratories, Abbott Park, Ill) was weight minutes was recorded as their “time to tices of community cardiologists in Cal- adjusted (40 mg/kg), with a maxi- ischemia.” Functional reserve was also · gary. Participants had to be aged 21 mum total dose for each treatment of measured by determination of VO2max years or older and have coronary ar- 3 g. Each treatment solution also con- and time to reach anaerobic threshold. tery disease proven by coronary angi- tained 750 mg of magnesium sulfate, Quality of life instruments included the ography or a documented myocardial 5 g of ascorbic acid, and5gofsodium Duke Activity Status Index,13 Health Sta- infarction and stable angina while re- bicarbonate (titrated to physiologic pH) tus Survey Short Form-36,14 and Se- ceiving optimal medical therapy. To in the 5% dextrose. Lidocaine, 80 mg, attle Angina Questionnaire.15 qualify for randomization patients were was added to relieve pain at the admin- required to have a treadmill test, us- istration site. In the placebo infusion so- Follow-up and Clinical Events ing a gradual ramping protocol, dem- lution the EDTA was replaced by 20 mL All patients were followed up for 1 year onstrating at least 1 mm of horizontal of 0.9% sodium chloride. The infu- from randomization. During this time, or downsloping ST-segment depression solutions were indistinguishable all clinical events were tabulated, includ- sion from the isoelectric line 80 milli- by color and labeling. The infusion so- ing death, myocardial infarction, coro- seconds after the J point. The study pro- lution was administered over 3 hours nary artery bypass graft surgery, and per- tocol required detection of ST- to minimize the potential unblinding cutaneous coronary intervention. segment depression between 2 and 14 effect of infusion-related adverse ef- All patients signed an informed con- minutes from the onset of exercise. fects. All patients received treatments sent form. The Conjoint Ethics Com- Exclusion criteria included planned twice weekly for 15 weeks and once mittee of the University of Calgary and revascularization, previous chelation monthly for an additional 3 months, for the Calgary Regional Health Author- therapy, evidence of heart failure, in- a total of 33 treatments. In accordance ity approved this study and its con- ability to walk on the treadmill, rest- with the ACAM protocol, patients in sent form. All clinical events were re- ing electrocardiographic (ECG) changes both groups took oral multivitamin ported to an independent safety that would interfere with ischemic as- therapy, 2 tablets 3 times daily as tol- monitoring committee. sessment, abnormal renal or liver func- erated, except on treatment days. All pa- tion, or untreated lipid abnormality at tients were seen at the University of Cal- Statistical Analysis the time of randomization. gary Cardiovascular Risk Reduction A sample size of 40 per group was cho- Clinic and had treatment of their risk sen to provide 90% power to detect a Treadmill Testing profile optimized according to Ameri- 60-second difference in mean change Treadmill testing was done at baseline can Heart Association guidelines (in- in exercise time from baseline to the 27- and at 15 and 27 weeks after random- cluding management of diet, lipid lev- week follow-up, assuming an SD of 80 ization. The protocol began with a level els, angina, stress, and exercise). seconds within each group. The 60- of exercise equivalent to 2 metabolic second difference was based on a mini- equivalents (METs) and increased Other Tests and Safety Monitoring mally important difference deter- slowly every 10 to 15 seconds, reach- Dipstick urine testing was performed mined by a consensus of Calgary ing an equivalent of 13 METs at 14 min- at each visit. Urinalysis and serum cre- cardiologists. Statistical analysis was

Downloaded from www.jama.com by guest on July 29, 2009 CHELATION THERAPY FOR ISCHEMIC HEART DISEASE performed using S-plus, version 6.0 Quality of Life ted at least once for worsening angina. (Mathsoft, Seattle, Wash). Categorical The changes in quality of life scores Four of these 7 patients had angio- variables were analyzed with the ␹2 or between baseline measurement and those plasty and none had coronary artery by- Fisher exact test, as appropriate. Con- obtained at the 27-week evaluation are pass graft (CABG) surgery for these tinuous variables were examined with shown in Table 2. There was a tendency events, although 1 other patient had paired and unpaired t tests. Graphical for modest increases in quality of life elective surgery (CABG was planned by examination of the data showed that scores in both groups with significant but the cardiologist after randomization normal assumption was viable. All re- similar improvements in the exertional without investigators’ knowledge). ported significance levels are 2-sided, capacity component of the Seattle Angina There was 1 myocardial infarction in and PϽ.05 was set as the significance Questionnaire. Differences between the the chelation group and 9 patients were level. All analyses of exercise and qual- groups were not significant. admitted at least once for worsening an- ity of life data were conducted using gina. None of these had angioplasty or last-observation-carried forward. Ischemia and Other Clinical Events CABG surgery associated with these Clinical events are presented on an in- events. RESULTS tention-to-treat basis (all 84 patients in- One of the chelation patients was Patients cluded). The duration of follow-up was withdrawn from therapy because of an A total of 3140 patients (FIGURE) were 1 year from randomization for each pa- elevation in serum creatinine. During screened and 171 of these agreed to un- tient. There were no deaths during that the first 10 treatments the patient’s se- dergo a qualifying exercise test. Eighty- time. One patient in the placebo group rum creatinine level increased from 1.5 four patients met the treadmill test cri- had a documented myocardial infarc- to 2.1 mg/dL (129 to 186 µmol/L, teria, consented, and were randomized tion and 6 other patients were admit- respectively). Treatment was stopped between January 1996 and January 2000. Baseline characteristics according to Figure. Flow of Patients in the Trial treatment assignment are shown in ABLE T 1. There were no important dif- 3140 Patients Screened ferences between the groups. Of the 84 patients randomized, 78 completed 2969 Excluded treatment, the final treadmill test, and 690 Refused Consent 576 Were Unable to Perform Treadmill/ECG the final quality of life assessments (39 470 Had Nonsignificant Disease in each group). Four placebo patients 412 Had Planned Revascularization 296 Had Comorbidity Factors and 2 chelation patients were unable to 137 Were Unavailable for Follow-up complete the treatment phase (Figure). 31 Had Previous EDTA Therapy 21 Died 336 Other Exercise End Points At baseline, mean (SD) treadmill test 171 Eligible for Treadmill Test times to ischemic ECG changes were 87 Did Not Meet Treadmill Test Criteria 572 (172) seconds in the placebo and 589 (176) seconds in the chelation 84 Randomized groups. Both groups were able to significantly (P<.001) increase their exercise time to ischemia at the 27-week 43 Assigned to Receive 41 Assigned to Receive treadmill test (TABLE 2). Changes in ex- Placebo Therapy EDTA Therapy ercise measurements of functional re- 4 Unable to Complete 2 Unable to Complete serve (time to anaerobic threshold and Protocol Protocol · 1 Cancer 1 Creatinine Elevated VO2max) are shown in Table 2. The 1 Pneumothorax Above Study Limit magnitude of the increases in time to 1 Unstable Angina 1 Hospitalization for ischemic changes and to anaerobic and Angioplasty Unstable Angina and 1 Coronary Artery Withdrawal of Consent threshold were not statistically differ- Bypass Graft Surgery ent in the 2 groups. The increase in · VO2max was not significant at the 27- 39 Completed Trial and 39 Completed Trial and week treadmill test in the placebo group Treadmill Test at Treadmill Test at 27 Weeks 27 Weeks but the increase in the chelation group was significant (P=.03). However, the 43 Included in Analysis 41 Included in Analysis difference between these 2 results was not significant (P=.46). ECG indicates electrocardiogram.

Downloaded from www.jama.com by guest on July 29, 2009 CHELATION THERAPY FOR ISCHEMIC HEART DISEASE and the serum creatinine level de- COMMENT chelation therapy involves multiple increased to 1.6 mg/dL (138 µmol/L) af- The main finding of this study was that fusions of EDTA to chelate lead, iron, ter 10 weeks. No other cause for the el- chelation therapy had no beneficial copper, calcium, and other metal ions evation in creatinine was found. In effect on exercise time to ischemia, in a redox inactive state. Chelated metal addition to the nonischemic events functional reserve for exercise, and ions are then excreted from the body shown in the Figure leading to discon- quality of life in patients with proven in the urine. For this reason EDTA has tinuation of therapy, 3 additional pla- ischemic heart disease, stable angina, been used as a chelating agent in clini- cebo patients were hospitalized for and evidence of ischemia on treadmill cal situations in which these elements nonischemic events: gout, lumbar back examination. Accordingly, chelation are found in excess.16 pain from a herniated disk, and gastro- therapy remains unproven in the treat- Because calcium is often found in intestinal bleeding. These events did not ment of ischemic heart disease. atheromatous plaques, early propo- interfere with completion of the treat- EDTA is an amino acid complex with nents hypothesized that EDTA might ment phase. There were no electrolyte a high affinity for divalent and triva- be effective in treating ischemic heart results out of normal range during the lent cations such as lead, magnesium, disease by liberating plaque calcium study. zinc, iron, and calcium. Conventional with a subsequent favorable change in the properties of the plaque.3,4,17,18 Other Table 1. Baseline Characteristics of Patients Completing Treatment* hypotheses possibly accounting for the Placebo Chelation symptomatic improvement reported by Variable (n = 43) (n = 41) P Value many patients with ischemic heart dis- Age, mean (SD), y 65 (8.5) 66 (9.1) .86 ease include a free radical–scavenging Male sex, No. (%) 32 (83.7) 33 (85.4) .93 function, inhibition of lipid oxidation Left ventricular ejection fraction, mean (SD), % 58 (13.1) 62 (11.2) .11 (antioxidant), reduction of total body Extent of CAD, No. (%) iron stores, cell membrane stabiliza- Single vessel 17 (39.5) 21 (51.2) .39 tion, arterial dilation due to possible cal- Multivessel 26 (60.5) 20 (48.8) cium channel blocking actions, or CCS angina class, No. (%) Asymptomatic 14 (32.6) 12 (29.3) stimulation of prostacyclin produc- I 17 (39.5) 22 (53.7) tion and improvement in arterial wall 3,4,17-21 II 9 (20.9) 5 (12.2) .53 elasticity. Oxidized cholesterol III 2 (4.7) 2 (4.9) plays an important role in endothelial IV 1 (2.3) 0 (0) function and the formation of athero- 22 Previous cardiac events, No. (%) sclerotic plaque. There is at least some Myocardial infarction 12 (27.9) 20 (48.8) .08 evidence, albeit controversial, that in- Percutaneous coronary intervention 19 (44.2) 19 (46.3) .98 creased total body iron stores are asso- Coronary artery bypass graft surgery 12 (27.9) 10 (24.4) .91 ciated with increased ischemic heart Comorbid conditions, No. (%) disease.23 Therefore, some of these hy- Diabetes 6 (14.0) 7 (17.1) .93 Hypertension 28 (65.1) 23 (56.1) .53 potheses about chelation having a po- Hyperlipidemia 34 (79.1) 34 (82.9) .86 tential mechanism for benefit in ische- Laboratory values, mean (SD), mg/dL† mic heart disease have plausibility. In Total cholesterol 185 (34.7) 185 (30.9) .77 the absence of studies confirming such HDL cholesterol 43 (8.9) 45 (13.1) .54 effects and, more importantly, confirm- LDL cholesterol 106 (27.0) 107 (22.4) .97 ing a definite clinical benefit of chela- Triglycerides 177 (132.7) 177 (79.6) .97 tion therapy, it remains possible that an- Creatinine 1.0 (0.20) 1.1 (0.23) .19 ecdotally reported improvements are Medication use, No. (%) simply due to the spontaneous fluctua- ␤ -Blockers 32 (74.4) 30 (73.2) .90 tions in symptoms frequently seen in Calcium channel blockers 23 (53.5) 19 (46.3) .51 ischemic heart disease.24-27 In our trial, Nitrates 19 (44.2) 10 (24.4) .06 the 1-minute increase in exercise time Triple therapy‡ 11 (25.6) 5 (12.2) .12 to ischemia and the improvement in the ACE inhibitors 13 (30.2) 11 (26.8) .73 exertional capacity component of the Aspirin 41 (95.3) 38 (92.7) .61 Seattle Angina Questionnaire in both Lipid-lowering agents 37 (86.0) 28 (68.3) .05 groups is consistent with a combina- *CAD indicates coronary artery disease; CCS, Canadian Cardiovascular Society; HDL, high-density lipoprotein; LDL, 24-27 low-density lipoprotein; and ACE, angiotensin-converting enzyme. The CCS scale is measured from I to IV, with a tion of placebo and training ef- higher score indicating greater severity. 28-30 †To convert cholesterol values to mmol/L, multiply values by 0.0259. To convert triglycerides to mmol/L, multiply val- fects commonly seen in studies of ues by 0.0113. To convert creatinine to µmol/L, multiply values by 88.4. angina patients. Another potential ex- ‡Triple therapy includes ␤-blockers, calcium channel blockers, and nitrates. planation for improvement is that both

Downloaded from www.jama.com by guest on July 29, 2009 CHELATION THERAPY FOR ISCHEMIC HEART DISEASE

Table 2. Exercise Times, Oxygen Consumption, and Quality of Life Scores at Baseline and 27 Weeks* Placebo Group Chelation Group Group Comparison

Baseline, 27 Weeks, Change, P Baseline, 27 Weeks, Change, P Difference in P Measurement Mean (SD) Mean (SD) Mean (95% CI) Value Mean (SD) Mean (SD) Mean (95% CI) Value Mean Change Value Time to ischemia, s 572 (172) 626 (186) 54 (23 to 84) Ͻ.001 589 (176) 652 (174) 63 (29 to 95) Ͻ.001 9 (−36 to 53) .69 Time to anaerobic 555 (151) 571 (195) 16 (−27 to 59) .45 555 (164) 585 (140) 31 (−11 to 72) .14 15 (−44 to 73) .63 threshold, s

V˙ o2max, mL/min 1606 (484) 1646 (419) 40 (−53 to 134) .39 1591 (468) 1675 (432) 84 (10 to 159) .03 44 (−74 to 162) .46 DASI 37.4 (13.4) 39.3 (14.5) 1.9 (−0.6 to 4.5) .13 42.2 (12.5) 41.9 (14.2) −0.2 (−3.2 to 2.7) .87 −2.1 (−6.0 to 1.6) .26 SAQ exertion 64.8 (20.3) 73.2 (17.8) 8.3 (3.9 to 12.8) Ͻ.001 69.9 (22.5) 77.2 (18.2) 7.3 (2.2 to 12.5) .006 −1.0 (−7.7 to 5.7) .77 SF-36 mental 48.3 (10.4) 50.4 (9.2) 2.1 (−0.4 to 4.5) .09 52.6 (7.6) 54.6 (6.7) 2.1 (−0.4 to 4.6) .10 0.01 (−3.4 to 3.4) .99 component summary SF-36 physical 39.9 (11.0) 44.9 (10.7) 5.0 (2.7 to 7.3) Ͻ.001 42.9 (10.1) 45.1 (10.0) 2.2 (−0.5 to 4.9) .11 −2.8 (−6.3 to 0.6) .11 component summary *CI indicates confidence interval. Maximum score on the Duke Activity Status Index (DASI) is 58.2, with a higher score indicating better physiologic reserve. Scores on the Seattle Angina Questionnaire (SAQ) range from 1-100, with a higher score indicating better levels of functioning. Scores on the Short-Form 36 (SF-36) range from 0-100, with a higher score indicating better health-related quality of life. Mean change values were rounded. groups were treated with optimal risk been criticized for the high dropout rate was no difference in the number of clini- reduction therapy. (123 completed 6-month follow-up). cal ischemic events, but our study was Chelation therapy is practiced and van Rij et al2 published the results of a not powered to detect any such differ- promoted as a form of complementary similar randomized, double-blind, pla- ences. According to our findings, the use or alternative medicine in many devel- cebo-controlled trial of walking time and of chelation therapy to increase ische- oped countries. Additional vitamins and ankle-brachial blood pressure indices in mic threshold and improve quality of life mineral supplements are recom- 32 patients (15 EDTA and 17 placebo) cannot be supported for patients with is- mended for patients undergoing che- with claudication, which also showed no chemic heart disease. Larger trials with lation therapy. In our study, both effect of EDTA therapy. a broader range of patients will be groups received multivitamins; we can- There is even less evidence in pa- needed to assess the safety and impact not exclude the possibility that these tients with ischemic heart disease. of EDTA therapy on clinical event rates. 5 supplements could be partially respon- Kitchell et al conducted a placebo- Author Contributions: Study concept and design: sible for the improvement that we saw controlled, double-blind, crossover Knudtson, Wyse, Galbraith, Brant, Hildebrand. in both groups. study of 9 patients with coronary heart Acquisition of data: Knudtson, Wyse, Galbraith, Hildebrand, Paterson, Richardson, Burkart, Burgess. In the literature, numerous authors disease and assessed performance on a Analysis and interpretation of data: Knudtson, Wyse, have reported positive results in uncon- treadmill. The authors documented that Galbraith, Brant, Burgess. 3-5,31 Drafting of the manuscript: Knudtson, Wyse, trolled studies. Very few random- 2 of 4 EDTA-treated patients ben- Galbraith, Brant, Hildebrand, Burgess. ized clinical trials have been published efited at 12 weeks but only 2 patients Critical revision of the manuscript for important in- tellectual content: Knudtson, Wyse, Galbraith, Brant, on the effects of chelation therapy, and volunteered to be treated in the sec- Hildebrand, Paterson, Richardson, Burkart. those that have been published were per- ond phase, and neither patient showed Statistical expertise: Wyse, Galbraith. formed in patients with peripheral ar- improvement. No statistical analyses Obtained funding: Knudtson, Wyse, Galbraith, Brant, Hildebrand. 1,2 32 terial disease. Olszewer et al pub- were presented in that study. Administrative, technical, or material support: Knudt- lished a small trial of 10 men with As with all randomized clinical tri- son, Wyse, Galbraith, Hildebrand, Paterson, Richard- son, Burgess. peripheral arterial disease in which im- als, our results can be applied to fit only Study supervision: Knudtson, Wyse, Galbraith, Pat- provement was demonstrated in walk- a similar population to that studied: pa- erson, Burkart, Burgess. Funding/Support: This study was supported by the ing distance after 20 treatments, but tients with stable angina who are not Alberta Health Services Research and Innovation Fund, there were only 5 patients in each group candidates for revascularization and can Medical Services Incorporated Research Foundation, and the Calgary Regional Health Authority. and therapy was not blinded. Guldager exercise on a treadmill. Our study PATCH Investigators: Clinical Steering Committee 1 and colleagues published a random- showed that following 33 treatments Members: Merril L. Knudtson, MD, D. George Wyse, ized, double-blind, placebo-controlled with EDTA therapy, there was no evi- MD, PhD, Rollin Brant, PhD, Ellen Burgess, MD, James Stone, MD, James Mayhew, MD, Jeanette Soriano, trial of 153 male patients with periph- dence of any benefit compared with pla- MD, Janet Hammond, P. Diane Galbraith, BN. Safety eral arterial disease (75 EDTA and 78 cebo in either objective measurements Monitoring Committee Members: Paul Armstrong, MD, University of Alberta; Koon Teo, MD, McMas- placebo), and during the 6-month fol- of exercise capacity or in measure- ter University; G. B. John Mancini, MD, University of low-up no effect of EDTA on walking ments of patient-perceived well-being. British Columbia. Substudy Committee: Merril L. Knudtson, MD, D. George Wyse, MD, PhD, Rollin time or ankle-brachial blood pressure in- One patient receiving EDTA had a tran- Brant, PhD, P. Diane Galbraith, BN, Todd Anderson, dex was demonstrated. That trial has sient increase in serum creatinine. There MD, Ellen Burgess, MD, and Derek Exner, MD, MPH.

Downloaded from www.jama.com by guest on July 29, 2009 CHELATION THERAPY FOR ISCHEMIC HEART DISEASE

Acknowledgment: We gratefully acknowledge David 10. Laine C, Venditti L, Localio R, Wickenheiser L, Mor- 22. Adams MR, Kinlay S, Blake GJ, Orford JL, Ganz Goodhart, MD, for his supervision during many ex- ris DL. Combined cardiac catheterization for uncom- P, Selwyn AP. Atherogenic lipids and endothelial dys- ercise tests. We thank Cliff Simpson for his computer plicated ischemic heart disease in a Medicare popu- function: mechanisms in the genesis of ischemic syn- programming expertise, the University of Calgary Lipid lation. Am J Med. 1998;105:373-379. dromes. Annu Rev Med. 2000;51:149-167. Clinic personnel, Heritage Medical Research Clinic per- 11. Rozema TC. Protocols for chelation therapy. J Adv 23. Salonen JT, Nyyssonen K, Korpela H, Tu- sonnel, the Foothills Medical Center pharmacy per- Med. 1997;10:1-100. omilehto J, Seppanen R, Salonen R. High stored iron sonnel, and Lu Mann, BN, for her help with the tread- 12. Ghali WA, Knudtson ML, for the APPROACH In- levels are associated with excess risk of myocardial in- mills. Joyce Forster helped with preparation of the vestigators. Overview of the Alberta Provincial Project farction in eastern Finnish men. Circulation. 1992;86: manuscript. for Outcome Assessment in Coronary Heart Disease. 803-811. Can J Cardiol. 2000;16:1225-1230. 24. Bienenfeld L, Frishman W, Glasser SP. The pla- 13. Hlatky MA, Boineau RE, Higginbotham MB, et al. cebo effect in cardiovascular disease. Am Heart J. 1996; REFERENCES A brief self-administered questionnaire to determine 132:1207-1221. 1. Guldager B, Jelnes R, Jorgensen SJ, et al. EDTA treat- functional capacity (the Duke Activity Status Index). 25. Weber JR. Problems associated with clinical evalu- ment of intermittent claudication: a double-blind, pla- Am J Cardiol. 1989;64:651-654. ation of antianginal medications. Am J Cardiol. 1985; cebo-controlled study. J Intern Med. 1992;231:261- 14. Ware JE Jr, Kosinski M, Keller SD. SF-36 Physical 56:14I-18I. 267. and Mental Health Summary Scales: A User’s Manual. 26. Benson H, McCallie DP Jr. Angina pectoris and 2. van Rij AM, Solomon C, Packer SG, Hopkins WG. Boston, Mass: Health Assessment Lab, New England the placebo effect. N Engl J Med. 1979;300:1424- Chelation therapy for intermittent claudication: a Medical Center; 1994. 1429. double-blind, randomized, controlled trial. Circula- 15. Spertus JA, Winder JA, Dewhurst TA, et al. De- 27. Folli G, Radice M, Beltrami A, Potenza S, Mari- tion. 1994;90:1194-1199. velopment and evaluation of the Seattle Angina Ques- otti G. Placebo effect in the treatment of angina pec- 3. Clarke NE, Clarke CN, Mosher RE. Treatment of tionnaire: a new functional status measure for coro- toris. Acta Cardiol. 1978;33:231-239. angina pectoris with disodium ethylene diamine tet- nary artery disease. J Am Coll Cardiol. 1995;25:333- 28. Romano M, Carella G, Cotecchia MR, et al. Ex- raacetic acid. Am J Med Sci. 1960;232:654-656. 341. ercise time: a possible source of misleading results dur- 4. Casdorph RH. EDTA chelation therapy: efficacy in 16. Dudley H, Ritchie A, Schilling A, Baker WH. Patho- ing long-term pharmacological studies by multiple stress atherosclerotic heart disease. J Hol Med. 1981;3:53- logical changes associated with the use of sodium eth- testings in coronary artery disease. Clin Cardiol. 1989; 59. ylene diamine tetraacetic acid. N Engl J Med. 1955; 12:39-41. 5. Kitchell JR, Palmon F, Aytan N, Meltzer L. The treat- 52:331-337. 29. Laslett LJ, Paumer L, Amsterdam EA. Increase ment of coronary artery disease with disodium EDTA. 17. Clarke NE, Clarke CN, Mosher RE. The “in vivo” in myocardial oxygen consumption indexes by Am J Cardiol. 1963;11:501-506. dissolution of metastatic calcium: an approach to ath- exercise training at onset of ischemia in patients 6. Ernst E. Chelation therapy for coronary heart dis- erosclerosis. Am J Med Sci. 1955;229:142-149. with coronary artery disease. Circulation. 1985;71: ease: an overview of all clinical investigations. Am Heart 18. Cranton EM, Frackelton JP. Free radical pathol- 958-962. J. 2000;140:139-141. ogy in age associated diseases: treatment with EDTA 30. Ades PA, Grunvald MH, Weiss RM, Hanson JS. 7. Eisenberg DM, Davis RB, Ettner SL, et al. Trends chelation, nutrition and antioxidants. J Hol Med. 1984; Usefulness of myocardial ischemia as predictor of train- in alternative medicine use in the United States, 1990- 6:6-37. ing effect in cardiac rehabilitation after acute myo- 1997: results of a follow-up national survey. JAMA. 19. Koen A, McCann D, Boyle A. Some in vivo ef- cardial infarction or coronary artery bypass grafting. 1998;280:1569-1575. fects of chelation-II: animal experimentation. J Chronic Am J Cardiol. 1989;63:1032-1036. 8. Grier MT, Meyers DG. So much writing, so little Dis. 1963;16:329-333. 31. Olszewer E, Carter JP. EDTA chelation therapy: science: a review of 37 years of literature on edetate 20. McCann D, Koen A, Zdybek G. Effect of chela- a retrospective study of 2,870 patients. J Adv Med. sodium chelation therapy. Ann Pharmacother. 1993; tion on phosphorus metabolism and experimental ath- 1989;2:197-211. 27:1504-1509. erosclerosis. Circ Res. 1962;9:880-884. 32. Olszewer E, Sabbag FC, Carter JP. A pilot double- 9. Quan H, Ghali WA, Verhoef MJ, Norris CM, Galbraith 21. Lamas GA, Ackermann A. Clinical evaluation of blind study of sodium-magnesium EDTA in periph- PD, Knudtson ML. Use of chelation therapy after coro- chelation therapy: is there any wheat amidst the chaff? eral vascular disease. J Natl Med Assoc. 1990;82:173- nary angiography. Am J Med. 2001;111:686-691. Am Heart J. 2000;140:4-5. 177.

Downloaded from www.jama.com by guest on July 29, 2009 Effect of Chelation Therapy on Endothelial Function in Patients With Coronary Artery Disease: PATCH Substudy

Todd J. Anderson,Jaroslav Hubacek,D.George Wyse, Merril L.Knudtson

Journal of the American College of Cardiology Vol. 41, No. 3, 2003 © 2003 by the American College of Cardiology Foundation ISSN 0735-1097/03/$30.00 Published by Elsevier Science Inc. doi:10.1016/S0735-1097(02)02770-5 Effect of Chelation Therapy on Endothelial Function in Patients With Coronary Artery Disease: PATCH Substudy Todd J. Anderson, MD, FRCP(C), Jaroslav Hubacek, MD, MSC, D. George Wyse, MD, PHD, FRCP(C), Merril L. Knudtson, MD, FRCP(C) Calgary, Canada

OBJECTIVES The purpose of this study was to evaluate the effect of chelation therapy with ethylenediamine tetraacetic acid (EDTA) on endothelium-dependent vasomotor responses in patients with documented coronary artery disease (CAD). BACKGROUND Oxidative stress plays an important role in the dysfunction of endothelium and development of atherosclerosis. Modification of cardiac risk factors and employment of antioxidants have been shown to improve endothelial function. Ethylenediamine tetraacetic acid chelation therapy is considered to be a complementary therapy for patients with CAD and is proposed to have antioxidant properties. METHODS A total of 47 patients enrolled in the Program to Assess Alternative Treatment Strategies to Achieve Cardiac Health (PATCH) participated in this substudy and had complete data. High-resolution ultrasound was used to assess endothelium-dependent brachial artery flow-mediated vasodilation (FMD) in patients with CAD in a randomized, double-blind, and placebo-controlled fashion. Patients were randomized to chelation therapy or placebo. The primary end point was the absolute difference in FMD after the first and 33rd treatments (6 months) of study groups compared with their baselines. RESULTS At the baseline, the study population had mild impairment of FMD (7.2 Ϯ 3.4%). The first chelation treatment did not change FMD as compared with placebo (chelation 6.5 Ϯ 3.5% vs. placebo 7.4 Ϯ 2.9%; p value ϭ 0.371). The brachial artery studies at six months did not demonstrate significant differences in FMD between study groups (placebo 7.3 Ϯ 3.4% vs. chelation 7.3 Ϯ 3.2%; p value ϭ 0.961). CONCLUSIONS Our results suggest that EDTA chelation therapy in combination with vitamins and minerals does not provide additional benefits on abnormal vasomotor responses in patients with CAD optimally treated with proven therapies for atherosclerotic risk factors. (J Am Coll Cardiol 2003;41:420–5) © 2003 by the American College of Cardiology Foundation

Chelation therapy or intravenous infusion of ethylenedia- important cause of ischemia in patients with established mine tetraacetic acid (EDTA) in combination with oral atherosclerosis (7). Furthermore, modification of cardiac vitamins and minerals is considered an “alternative” or risk factors and employment of antioxidants have been “complementary” therapy for patients with coronary artery shown to improve endothelial function (8). On the basis of disease (CAD). It has been suggested that patients with previously proposed antioxidant, antihypertensive, and CAD might benefit from this complementary therapeutic cholesterol-lowering properties of chelation therapy (9), we approach (1–4). However, definitive evidence that chelation hypothesized that EDTA in combination with vitamins and is beneficial has been lacking in randomized trials. Several minerals may have a beneficial effect on endothelial function mechanisms for the potential beneficial effects of chelation in patients with established CAD. therapy in patients with atherosclerosis have been postu- The purpose of this study was to evaluate the effect of lated. Lamas and Ackermann (5) suggested that an antiox- chelation therapy with EDTA on endothelium-dependent idant effect of EDTA, in combination with vitamins and vasomotor responses in patients with documented CAD in minerals, might be the most plausible. a randomized, double-blind, and placebo-controlled fash- Over the past decade, emerging data have demonstrated ion. that cardiovascular risk factors and oxidative stress play a crucial role in abnormal vasomotor responses (6). Abnormal METHODS vasomotor responses are probably one of the first events in the atherosclerosis process and have been shown to be an Study subjects. A total of 53 patients enrolled in the Program to Assess Alternative Treatment Strategies to Achieve Cardiac Health (PATCH) study participated in From the Faculty of Medicine, University of Calgary and Calgary Health Region, Calgary, Canada. Supported by the Alberta Health Services Research and Innovation this substudy. The main PATCH study has been previously Fund, the Medical Services Incorporated Research Foundation, and the Calgary reported (10). Briefly, patients Ն21 years old with CAD Regional Health Authority. Todd J. Anderson is a Medical Scholar of the Alberta proven by coronary angiography or a documented myocar- Heritage Foundation for Medical Research. Manuscript received June 7, 2002; revised manuscript received October 2, 2002, dial infarction and stable angina on optimal medical therapy accepted October 17, 2002. were studied. To qualify for randomization, patients were JACC Vol. 41, No. 3, 2003 Anderson et al. 421 February 5, 2003:420–5 Effect of EDTA Chelation on Endothelial Function

took oral multivitamin therapy (FLW, Douglas Labs, Pitts- Abbreviations and Acronyms burgh, Pennsylvania), two tablets three times daily as ACE ϭ angiotensin-converting enzyme tolerated, except on treatment days. The components of CAD ϭ coronary artery disease FLW were as follows: vitamin A 4,000 IU, vitamin E 65 ϭ EDTA ethylenediamine tetraacetic acid IU, vitamin C 400 mg, vitamin B1 20 mg, vitamin B2 5 mg, eNOS ϭ endothelial nitric oxide synthase ␮ FMD ϭ flow-mediated vasodilation vitamin B6 15 mg, vitamin B12 25 g, niacin 5 mg, NO ϭ nitric oxide niacinamide 5 mg, pantothenic acid 50 mg, folic acid 0.04 NTG ϭ nitroglycerin mg, biotin 10 ␮g, choline 72.5 mg, inositol 5 mg, methio- PATCH ϭ Program to Assess Alternative Treatment nine 24 mg, magnesium 40 mg, potassium 40 mg, manga- Strategies to Achieve Cardiac Health nese 0.5 mg, zinc 3 mg, chromium 20 ␮g, and selenium 25 ␮g. All patients were seen at the University of Calgary Cardiovascular Risk Reduction Clinic and had treatment of their risk profile optimized. required to have a treadmill test using a gradual ramping BRACHIAL ENDOTHELIAL FUNCTION STUDIES. A recently protocol and demonstrating Ն1 mm horizontal or described technique employing high-resolution ultrasound downsloping ST depression from the isoelectric line 80 ms to measure brachial artery diameter was used to study the after the J point. The study required detection of ST vasodilator responses induced by reactive hyperemia after 5 depression between 2 and 14 min from the onset of exercise. min of upper arm occlusion (11). Assessment of brachial Patients were excluded for the following reasons: planned artery endothelial function was performed at baseline, after revascularization, previous chelation therapy, evidence of the first chelation therapy (acute), and at the end of study heart failure, inability to walk on the treadmill, resting (after the completion of 33 chelation treatments) by a electrocardiogram changes that would interfere with isch- previously validated technique (12,13). Patients underwent emic assessment, abnormal renal or liver function, or un- each of three brachial artery ultrasound studies after an treated lipid abnormality at the time of randomization. overnight fast (12 h), and all vasoactive medications, includ- Study protocol. RANDOMIZATION AND TREATMENT. ing angiotensin-converting enzyme (ACE) inhibitors, beta- Written informed consent was obtained from all partici- blockers, calcium-channel blockers, long-acting nitrites, and pants. The Conjoint Ethics Committee of the University of statins were stopped for 24 h. Studies were performed in a Calgary and the Calgary Regional Health Authority ap- quiet clinical laboratory with the temperature maintained at proved this study and its consent form. All clinical events 21 to 23°C. A 7.5 MHz linear phase-arrayed ultrasound were reported to an independent Safety Monitoring Com- transducer attached to a Hewlett-Packard 5500 ultrasound mittee. The present study was a double-blind, randomized, machine was used. Pulsed-wave Doppler was used to record and placebo-controlled trial. Patients were randomized in brachial artery velocity for each of the interventions. blocks of 10 using S-plus version 3.4 (Statistical Science Data analysis. BRACHIAL ARTERY ANALYSIS. Brachial ar- Inc., Seattle, Washington). The Foothills Medical Center tery analysis was performed at a core laboratory at the Pharmacy assigned the randomized therapy and prepared University of Calgary by a single technician blind to the solutions for blinded administration of infusions. The 500 study group assignment or study sequence. Three sequential ml infusion solution of 5% dextrose in water for the active systolic frames (taken at the end of the T-wave on the treatment containing disodium EDTA (Endrate, Abbott electrocardiogram) were digitized via an analog-to-digital Laboratories) was weight adjusted (40 mg/kg), with a converting board. A software algorithm automatically cal- maximum total dose for each treatment of 3 g. Each culates the average diameter (100 points) over the operator- treatment solution also contained 750 mg of magnesium selected segment. Flow-mediated vasodilation (FMD) was sulfate,5gofascorbic acid (Mega C), and 5 g sodium calculated from the diameters as (reactive hyperemia Ϫ bicarbonate (titrated to physiologic pH) in the D W. 5 baseline)/baseline ϫ 100%. The intra- and inter-observer Lidocaine (Xylocaine ) 80 mg was added to relieve pain at variability in our laboratory is 1%. Systolic frames were used the administration site and mask taste. In the placebo as previously validated (13). In 20 studies selected randomly, infusion solution, the EDTA was replaced by 20 ml of 0.9% measurements were made at both end-diastole and systole, sodium chloride. The infusion solutions were indistinguish- and the FMD was exactly the same. Brachial artery flow was able by color, taste, and labeling. All infusion solutions were calculated as the product of velocity and cross-sectional prepared following manufacturers’ instructions and were arterial area. administered immediately after mixing. The infusion solution was administered over 3 h to minimize the potential STATISTICS. Data are expressed as mean Ϯ SD for contin- unblinding effect of infusion-related side effects. All patients uous variables and as counts and percentage for discrete received treatments twice weekly for 15 weeks and once per variables. Statistical analyses were conducted with a com- month for an additional three months, for a total of 33 mercially available software package (SPSS version 10.0; treatments. In accordance with the American College of SPSS Inc, Chicago, Illinois). The sample size was deter- Advancement in Medicine protocol, patients in both groups mined to detect a clinically important absolute improvement 422 Anderson et al. JACC Vol. 41, No. 3, 2003 Effect of EDTA Chelation on Endothelial Function February 5, 2003:420–5

Table 1. Baseline Demographics Table 2. Brachial Artery Parameters Placebo Chelation Placebo Chelation p Value p Value (24 ؍ n) (23 ؍ n) Baseline Age 64 Ϯ 964Ϯ 9 0.939* Baseline diameter (mm) 4.1 Ϯ 0.7 4.0 Ϯ 0.8 0.974 Male 19 (83%) 21 (88%) 0.641 FMD (%) 7.6 Ϯ 3.4 6.7 Ϯ 3.5 0.355 Female 4 (17%) 3 (13%) 0.641 NTG (%) 12.8 Ϯ 5.0 13.5 Ϯ 6.3 0.664 Postmenopausal 4 (17%) 3 (13%) 0.641 Short-term Diabetes mellitus 1 (4%) 4 (17%) 0.176 Baseline diameter (mm) 4.1 Ϯ 0.8 4.1 Ϯ 0.7 0.769 Hypertension 11 (48%) 9 (38%) 0.479 FMD (%) 7.4 Ϯ 2.9 6.5 Ϯ 3.5 0.371 Smoker 1 (4%) 1 (4%) 0.976 NTG (%) 12.0 Ϯ 4.9 12.8 Ϯ 4.1 0.516 CHF 1 (4%) 1 (4%) 0.976 6-month PVD 3 (13%) 2 (8%) 0.605 Baseline diameter (mm) 4.1 Ϯ 0.5 4.0 Ϯ 0.8 0.625 Lipids (mmol/l) FMD (%) 7.3 Ϯ 3.4 7.3 Ϯ 3.2 0.961 Total cholesterol 5.0 Ϯ 1.1 4.9 Ϯ 0.8 0.760* NTG (%) 12.4 Ϯ 6.4 15.3 Ϯ 7.1 0.148 LDL-C 3.0 Ϯ 0.7 2.9 Ϯ 0.7 0.639* baseline ϭ measurement performed before the treatment, after the ﬁrst chelation HDL-C 1.1 Ϯ 0.2 1.1 Ϯ 0.3 0.975* Ϯ Ϯ therapy (short-term) and after the completion of 33 chelation treatments; data Triglycerides 2.0 1.1 2.8 3.9 0.356* presented in millimeters as mean Ϯ SD for baseline diameter and in % Ϯ SD increase from baseline diameter for FMD and NTG; FMD ϭ ﬂow-mediated vasodilation; CHF ϭ history of congestive heart failure; HDL-C ϭ high-density lipoprotein ϭ ϭ ϭ ϭ NTG nitroglycerin-mediated vasodilation; p value chelation therapy versus cholesterol; LDL-C low-density lipoprotein cholesterol; p value chelation placebo group; unpaired t test. therapy versus placebo group; unpaired t test and Mann-Whitney U-test (*); PVD ϭ peripheral vascular disease

aspirin, 80% with lipid-lowering therapies, 70% with beta- in FMD of 3% (with a SD of 3% to 4%) in the chelation blockers, 34% with ACE inhibitors, and 30% with long- group when compared with placebo. This assumes an alpha acting nitrates. The use of medication was comparable of 0.05 and beta 0.2. Categorical variables were analyzed between the randomization groups. During the course of with the Mann-Whitney U-test. Continuous variables were the study, there was no statistically significant difference in examined with paired and unpaired t tests, as appropriate. lipid profile or in blood pressure between the groups. All reported significance levels are two-sided. Treatment effect on brachial artery characteristics. Flow- mediated vasodilation was inversely related to baseline RESULTS brachial artery diameter in this study (baseline 4.1 Ϯ 0.7 mm vs. FMD diameter 4.3 Ϯ 0.8 mm; p value Ͻ0.001) and Patient characteristics. A total of 53 patients agreed and in other studies (13–15). The baseline brachial artery consented to participate in the study between January 1996 diameter was the same for both study groups, and the study and January 2000. Of the 53 patients, 47 completed medications had no significant effect on baseline brachial treatment and final brachial artery ultrasound study. Four diameter after six months of treatment. Upper arm occlu- placebo patients were unable to finish the treatment phase sion resulted in an increase in forearm blood flow of because of: 1) cancer of the hip; 2) hospitalization with approximately 450%, which was the same at baseline for pneumothorax in the setting of previous asbestosis and both groups and was not affected by drug treatment. chronic obstructive pulmonary disease; 3) unstable angina Treatment effect on FMD. At the baseline, the study followed by angioplasty and coronary artery bypass graft population had FMD of 7.2 Ϯ 3.4% and nitroglycerin surgery; and 4) coronary artery bypass graft surgery that had Ϯ been planned as an elective procedure by the primary (NTG)-mediated dilation of 13.2 5.7%. These values are cardiologist after randomization without the knowledge of similar to what we have reported previously in patients with the investigators. Two EDTA patients were unable to CAD (14) and less than those of normal controls in our lab complete the study: one because of a serum creatinine level (16). Baseline FMD and NTG-mediated vasodilations were increase above preset study limit (175 ␮mol/l) and the other similar in both groups (Table 2). Short-term treatment with because of hospitalization for unstable angina followed by EDTA in combination with vitamins and minerals did not withdrawal of consent. There was no difference in the changed FMD and NTG-mediated vasodilation (Table 2). demographics of these patients and those who completed The brachial artery studies at six months, after the comple- the study. Data presented in this report are from the 47 tion of 33 treatments, did not demonstrate significant patients with complete information. The baseline charac- differences in FMD between placebo and chelation therapy teristics of the 47 randomized patients are shown in Table 1. groups (placebo 7.3 Ϯ 3.4% vs. chelation 7.3 Ϯ 3.2%; p There were no important differences at baseline between the value ϭ 0.961) (Table 2). There was no significant effect of two groups. The mean age of the study population was 68 chelation therapy on NTG-induced vasodilation (placebo Ϯ 9 years, with 85% men. Risk factors were as follows: 12.4 Ϯ 6.4% vs. chelation group 15.3 Ϯ 7.1%; p value ϭ hypercholesterolemia 80%, hypertension 43%, and diabetes 0.148) (Table 2). Furthermore, we did not observe any mellitus 10%. All patients had treatment of their risk profile statistically significant changes within either study group optimized. Among all 47 patients, 95% were treated with over time (Fig. 1). JACC Vol. 41, No. 3, 2003 Anderson et al. 423 February 5, 2003:420–5 Effect of EDTA Chelation on Endothelial Function

Figure 1. Changes in vasomotor responses in study groups over time. (A) placebo group; (B) chelation therapy group; *p value Ͻ 0.05 (baseline vs. acute vs. 6 months) paired t test. FMD ϭ ﬂow-mediated vasodilation; NTG ϭ nitroglycerin.

DISCUSSION in coronary and peripheral circulation (12,18). Oxygen free radicals, generated by a number of pathways in the body, This randomized, double-blind, and placebo-controlled play an important role in the dysfunction of endothelium study demonstrates that short- and long-term treatment (33 and development of atherosclerosis (19). Modification of infusions) with EDTA in combination with vitamins and minerals did not improve impaired endothelium-dependent cardiac risk factors and employment of antioxidants have FMD in the brachial artery circulation of patients with been shown to improve endothelial function (8). established CAD and optimized treatment of risk profile. Chelation therapy is considered to be complementary There is ample evidence that the endothelium plays an therapy for patients with CAD. In fact, the beneficial effect important role in maintaining vascular integrity through the of EDTA therapy in this setting remains unproven (5,10). release of a variety of paracrine and autocrine factors, Originally, liberation of plaque calcium with a subsequent including nitric oxide (NO). In health, NO causes vasodi- favorable change in the properties of the plaque (20–23) was lation, inhibits platelet aggregation and adhesion, and in- thought to be the underlying mechanism of action. Later, hibits smooth muscle cell proliferation and white cell EDTA was proposed to lower low-density lipoprotein and adhesion (17). Endothelial dysfunction has been observed in very-low-density lipoprotein levels and iron stores, inhibit patients with established CAD or coronary risk factors, both platelet aggregation, relax vascular tone, and “scavenge” free 424 Anderson et al. JACC Vol. 41, No. 3, 2003 Effect of EDTA Chelation on Endothelial Function February 5, 2003:420–5 radicals (5,20–29). Ethylenediamine tetraacetic acid reduces vasomotor responses in humans (47–50). Therefore, it can iron and copper levels from cell membrane. Both these be speculated that free radical production was suppressed metals are important catalysts in the peroxidation of unsat- below the level at which free radical scavengers used in the urated fatty acids and the oxidation of low-density lipopro- presented study (EDTA and vitamin C) could demonstrate tein, which generate free radicals with subsequent disrup- their beneficial effect on vasomotor responses. It was a tion of membrane architecture, promoting cellular injury limitation of our study that we did not have a measure of and progression of atherosclerosis (30). The beneficial effect oxidative stress or vitamin C levels. of iron chelation on endothelial function has been previously In conclusion, our results suggest that chelation therapy demonstrated. Deferoxamine, an iron chelator, improved with EDTA in combination with vitamins and minerals NO-mediated, endothelium-dependent vasodilation in pa- does not provide any additional benefit on abnormal vaso- tients with CAD (31) and in diabetics with angiographically motor responses in patients with CAD optimally treated for normal coronary arteries (32). On the other hand, in a atherosclerotic risk factors with proven therapies. non-randomized and non-blinded study (33), EDTA alone did not significantly improve either endothelium or NO- Reprint requests and correspondence: Dr. Todd J. Anderson, dependent vasodilation in patients with CAD after 10 1403 29th Street NW, Calgary, Alberta T2N 2T9, Canada. chelation infusions over 60 days. However, EDTA in E-mail: [email protected]. combination with predominantly B vitamins improved vasomotor responses and was associated with a significant REFERENCES reduction in homocysteine levels. The patients in our study 1. Holden C. Alternative medicine: NIH trial to test chelation therapy. had documented CAD and impaired endothelial function in Science 2002;297:1109. peripheral circulation at baseline. An improvement in ab- 2. Quan H, Ghali WA, Verhoef M, Norris CM, Galbraith PD, Knudtson ML. Use of chelation therapy after coronary angiography. normal vasomotor responses has not been observed in Am J Med 2001;111:686–91. patients randomized to EDTA therapy compared with 3. Prozema TC. Protocols for chelation therapy. J Adv Med 1997;10:1– patients randomized to placebo, either in the short term or 100. 4. Cappel LT, Janson M. EDTA chelation therapy in the treatment of at six months. Furthermore, we have not demonstrated any vascular disease. J Cardiovasc Nursing 1996;10:78–86. statistically significant changes in FMD or NTG-mediated 5. Lamas GA, Ackermann A. Clinical evaluation of chelation therapy: is vasodilation within the study groups over time. One possi- there any wheat amidst the chaff? Am Heart J 2000;140:4–5. 6. Anderson TJ. Oxidative stress, endothelial function and coronary ble explanation might be that EDTA, unlike deferoxamine, atherosclerosis. Cardiologia 1997;42:701–14. is too small to bind entire iron ions and it creates a seventh 7. Benzuly KH, Padgett RC, Kaul S, Piegors DJ, Armstrong ML, coordination site (34). As a result, iron remains in its Heistad DD. Functional improvement precedes structural regression of atherosclerosis. Circulation 1994;89:1810–8. catalytically active form, and EDTA-iron complex can 8. Hornig B, Drexler H. Reversal of endothelial dysfunction in humans. catalyze the formation of the hydroxyl radical in a superox- Coron Artery Dis 2001;12:463–73. ide generating system (35). 9. Elihu N, Anandasbapathy S, Frishman WH. Chelation therapy in cardiovascular disease: enthylendiaminetetraacetic acid and desferox- In our study, “placebo treatment” contained a high dose amine. J Clin Pharmacol 1998;38:101–5. of intravenous vitamin C and multivitamin oral supplemen- 10. Knudtson M, Galbraith PD. Chelation therapy for ischemic heart tation. Therefore, we cannot exclude a beneficial effect of disease. A randomized controled trial. JAMA 2002;287:481–6. 11. Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive treatment with vitamins and minerals alone. Vitamin C is a detection of endothelial dysfunction in children and adults at risk of water-soluble antioxidant that inhibits lipid peroxidation in atherosclerosis. Lancet 1992;340:1111–5. vitro (36) and in vivo (37) by directly scavenging aqueous 12. Anderson TJ, Uehata A, Gerhard MD, et al. Close relation of endothelial function in the human coronary and peripheral circula- free radicals, by reducing the chain-carrying alpha- tions. J Am Coll Cardiol 1995;26:1235–41. tocopheroxyl radical, and by stabilization of tetrahydrobiop- 13. Sorensen KE, Celermajer DS, Spiegelhalter DJ, et al. Non-invasive terin, which prevents uncoupling of eNOS (38,39). In measurement of human endothelium dependent arterial responses: accuracy and reproducibility. Br Heart J 1995;74:247–53. clinical studies, vitamin C has been shown to reverse 14. Anderson TJ, Elstein E, Haber HE, Charbonneau F. Comparative impaired endothelial function in patients with established study of ACE-inhibition, angiotensin II antagonism, and calcium CAD (40,41). However, long-term vitamin C supplemen- channel blockade on flow-mediated vasodilation in patients with coronary disease (BANFF study). J Am Coll Cardiol 2000;35:60–6. tation did not show a consistent beneficial effect on the 15. Celermajer DS, Sorensen KE, Gooch VM, et al. Non-invasive endothelial function of treated patients (42). The reason detection of endothelial dysfunction in children and adults at risk of why our patient population, either in the EDTA or placebo atherosclerosis. Lancet 1992;340:1111–5. 16. Schnell G, Robertson A, Malley L, Huston D, Anderson TJ. Impaired group, did not respond to vitamin C treatment is not clear. brachial artery endothelial function is not predicted by elevated All our patients were aggressively treated for atherosclerotic triglycerides. J Am Coll Cardiol 1999;33:2038–43. risk factors. The majority of our study population was 17. Moncada S, Higgs A. The L-arginine-NO pathway. N Engl J Med 1993;329:2002–12. treated with lipid-lowering therapy and 34% of patients 18. Drexler H. Endothelial dysfunction: clinical implications. Prog Car- used ACE inhibitors. Lipid-lowering therapies and ACE diovasc Dis 1997;39:287–324. inhibition have been shown to augment endothelial NO 19. Anderson TJ. Oxidative stress, endothelial function and coronary atherosclerosis. Cardiologia 1997;42:701–14. synthesis, to inhibit endothelial superoxide anion in animal 20. Casdorph RH. EDTA chelation therapy: efficacy in atherosclerotic studies (43–46), and to have beneficial effects on abnormal heart disease. J Holistic Med 1981;3:53–9. JACC Vol. 41, No. 3, 2003 Anderson et al. 425 February 5, 2003:420–5 Effect of EDTA Chelation on Endothelial Function

21. Clarke NE, Clarke CN, Mosher RE. The “in vivo” dissolution of 38. Heller R, Unbehaun A, Schellenberg B, Mayer B, Werner-Felmayer metastic calcium: an approach to atherosclerosis. Am J Med Sci G, Wermer ER. L-ascorbic acid potentiates endothelial nitric oxide 1955;229:142–9. synthesis via a chemical stabilization of tetrahydrobiopterin. J Biol 22. Clarke NE, Clarke CN, Mosher RE. Treatment of angina pectoris Chem 2001;276:40–7. with disodium ethylene diamine tetraacetic acid. Am J Med Sci 39. Tiefenbacher CP. Tetrahydrobiopterin: a critical cofactor for eNOS 1960;232:654–6. and a strategy in the treatment of endothelial dysfunction. Am J 23. Cranton EM, Frackelton JP. Free radical pathology in age associated Physiol 2001;280:H2484–8. diseases: treatment with EDTA chelation, nutrition and antioxidants. 40. Gokce N, Keaney JFJ, Frei B, et al. Long-term ascorbic acid J Holistic Med 1984;6:6–37. administration reverses endothelial vasomotor dysfunction in patients 24. Frishman WH. Chelation therapy for coronary artery disease: panacea with coronary artery disease. Circulation 1999;99:3234–40. or quackery? Am J Med 2001;111:729–30. 41. Levine GN, Frei B, Koulouris SN, Gerhard MD, Keaney JF Jr, Vita 25. Kindness G, Frackelton JP. Effect of ethylendiaminetetraacetic acid JA. Ascorbic acid reverses endothelial vasomotor dysfunction in (EDTA) on platelet aggregation in human blood. J Adv Med patients with coronary artery disease. Circulation 1996;93:1107–13. 1989;2:519–29. 42. Raitakari O, Adams MR, McCredie R, Griffiths KA, Stocker R, 26. Koen A, McCann D, Boyle A. Some in vivo effects of chelation. II. Celermajer DS. Oral vitamin C and endothelial function in smokers: Animal experimentation. J Chronic Dis 1963;16:329–33. short-term improvement but no sustained beneficial effect. J Am Coll 27. McCann D, Koen A, Zdybek G. Effect of chelation on phosphorus Cardiol 2000;35:1616–21. metabolism and experimental atherosclerosis. Circ Res 1962;9:880–4. 43. Rajagopalan S, Kurz S, Munzel T, et al. Angiotensin II-mediated 28. Olszewer E, Carter JP. EDTA chelation therapy in chronic degener- hypertension in the rat increases vascular superoxide production via ative disease. Med Hypotheses 1988;27:41–9. membrane NADH/NADPH oxidase activation. Contribution to al- 29. Olszewer E, Sabbag FC, Carter JP. A pilot double-blind study of terations of vasomotor tone. J Clin Invest 1996;97:1916–23. sodium-magnesium EDTA in peripheral vascular disease. J Natl Med 44. Wagner AH, Kohler T, Ruckschloss U, Just I, Hecker M. Improve- Assoc 1990;82:173–7. ment of nitric oxide-dependent vasodilatation by HMG-CoA reduc- 30. Deucher DP. EDTA chelation: an antioxidant strategy. J Adv Med tase inhibitors through attenuation of endothelial superoxide anion 1988;1:182–90. formation. Arterioscler Thromb Vasc Biol 2000;20:61–9. 31. Duffy SJ, Biegelsen ES, Holbrook M, et al. Iron chelation improves 45. Warnholtz A, Nickenig G, Schulz E, et al. Increased NADH-oxidase- endothelial function in patients with coronary artery disease. Circula- tion 2001;103:2799–804. mediated superoxide production in the early stages of atherosclerosis: 32. Nitenberg A, Payacha F, Ledoux S, Sachs R, Attali J-R, Valensi P. evidence for involvement of the renin-angiotensin system. Circulation Coronary artery responses to physiological stimuli are improved by 1999;99:2027–33. deferoxamine but not by L-arginine in non-insulin-dependent diabetic 46. Wassmann S, Laufs U, Baumer AT. Inhibition of geranylgeranylation patients with angiographically normal coronary arteries and no other reduces angiotensin II-mediated free radical production in vascular risk factors. Circulation 1998;97:736–43. smooth muscle cells: involvement of angiotensin AT1 receptor expres- 33. Green DJ, O’Driscoll JG, Maiorana A, Scrimgeour NB, Weerasooriya sion and Rac1 GT phase. Mol Pharmacol 2001;59:646–54. R, Taylor RR. Effect of chelation with EDTA and vitamin B therapy 47. Anderson TJ, Meredith IT, Yeung AC, Frei B, Selwyn AP, Ganz P. on nitric oxide-related endothelial vasodilator funstion. Clin Exp The effect of cholesterol-lowering and antioxidant therapy on Pharmacol Physiol 1999;26:853–6. endothelium-dependent coronary vasomotion. N Engl J Med 1995; 34. Marx JJM, Vreugdenhil MD, Voest EE. Iron chelating agents are not 332:488–93. useful in treating atherosclerosis (letter). Ann Intern Med 1994;121. 48. Mancini GBJ, Henry GC, Macaya C, et al. Angiotensin converting 35. McCord JM, Day ED Jr. Superoxide-dependent production of hy- enzyme inhibition with quinapril improves endothelial vasomotor droxyl radical catalyzed by iron-EDTA complex. FEBS Lett 1978;86: dysfunction in patients with coronary artery disease: the TREND 139–42. study. Circulation 1996;94:258–65. 36. Carr AC, Zhu BZ, Frei B. Potential antiatherogenic mechanisms of 49. Stroes ES, Koomans HA, de Bruin TW, Rabelink TJ. Vascular ascorbate (vitamin C) and alpha-tocopherol (vitamin E). Circ Res function in the forearm of hypercholesterolaemic patients off and on 2000;87:349–54. lipid-lowering medication. Lancet;1995;19;346;467–71. 37. Chen K, Suh J, Carr AC. Vitamin C suppresses oxidative lipid damage 50. Treasure CB, Klein L, Weintraub WS, et al. Beneficial effects of in vivo, even in the presence of iron overload. Am J Physiol Endocrinol cholesterol lowering therapy on the coronary endothelium in patients Metab 2000;279:E1406–12. with coronary artery disease. N Engl J Med 1995;332:481–7. Chelation therapy for atherosclerotic cardiovascular disease (The Cochrane Review)

Dans AL,Tan FN,Villarruz-Sulit EC

Chelation therapy for atherosclerotic cardiovascular disease (Review)

Dans AL, Tan FN, Villarruz-Sulit EC

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2009, Issue 3 http://www.thecochranelibrary.com

HEADER...... 1 ABSTRACT ...... 1 PLAINLANGUAGESUMMARY ...... 2 BACKGROUND ...... 2 OBJECTIVES ...... 2 METHODS ...... 2 RESULTS...... 3 DISCUSSION ...... 5 AUTHORS’CONCLUSIONS ...... 5 ACKNOWLEDGEMENTS ...... 5 REFERENCES ...... 5 CHARACTERISTICSOFSTUDIES ...... 6 DATAANDANALYSES...... 13 Analysis 1.1. Comparison 1 EDTA versus placebo, Outcome 1 Ankle-brachial pressure index at 20th infusion. . . 14 Analysis 1.2. Comparison 1 EDTA versus placebo, Outcome 2 Ankle-brachial pressure index at 3 months post-treatment. 15 Analysis 1.3. Comparison 1 EDTA versus placebo, Outcome 3 Ankle-brachial pressure index at 6 months post-treatment. 15 Analysis 1.4. Comparison 1 EDTA versus placebo, Outcome 4 Walking distance (m) at 10th infusion...... 16 Analysis 1.5. Comparison 1 EDTA versus placebo, Outcome 5 Walking distance (m) at 20th infusion...... 16 Analysis 1.6. Comparison 1 EDTA versus placebo, Outcome 6 Walking distance (m) at 3 months post-treatment. . 17 Analysis 1.7. Comparison 1 EDTA versus placebo, Outcome 7 Walking distance (m) at 6 months post-treatment. . 17 Analysis 1.8. Comparison 1 EDTA versus placebo, Outcome 8 Pain-free walking distance (m) at 10th infusion. . . 18 Analysis 1.9. Comparison 1 EDTA versus placebo, Outcome 9 Pain-free walking distance (m) at 20th infusion. . . 18 Analysis 1.10. Comparison 1 EDTA versus placebo, Outcome 10 Pain-free walking distance (m) at 3 months post- treatment...... 19 Analysis 1.11. Comparison 1 EDTA versus placebo, Outcome 11 Pain-free walking distance (m) at 6 months post- treatment...... 19 Analysis 1.12. Comparison 1 EDTA versus placebo, Outcome 12 Subjective walking distance (m) at 20th infusion. . 20 Analysis 1.13. Comparison 1 EDTA versus placebo, Outcome 13 Subjective walking distance (m) at 3 months post- treatment...... 20 Analysis 1.14. Comparison 1 EDTA versus placebo, Outcome 14 Subjective evaluation - aggravation...... 21 Analysis 1.15. Comparison 1 EDTA versus placebo, Outcome 15 Subjective evaluation - unchanged...... 21 Analysis 1.16. Comparison 1 EDTA versus placebo, Outcome 16 Subjective evaluation - mild improvement. . . . 22 Analysis 1.17. Comparison 1 EDTA versus placebo, Outcome 17 Subjective evaluation - moderate improvement. . 22 Analysis 1.18. Comparison 1 EDTA versus placebo, Outcome 18 Subjective evaluation - great improvement. . . . 23 Analysis 1.19. Comparison 1 EDTA versus placebo, Outcome 19 Plasma low density lipoprotein (LDL)-cholesterol levels (mmol/litre)...... 23 Analysis 1.20. Comparison 1 EDTA versus placebo, Outcome 20 Plasma triglyceride levels (mmol/litre). . . . . 24 Analysis 1.21. Comparison 1 EDTA versus placebo, Outcome 21 Plasma high density lipoprotein (HDL)-cholesterol levels (mmol/litre)...... 24 Analysis 1.22. Comparison 1 EDTA versus placebo, Outcome 22 Plasma cholesterol levels (mmol/litre)...... 25 Analysis 1.23. Comparison 1 EDTA versus placebo, Outcome 23 Adverse event - stroke...... 25 Analysis 1.24. Comparison 1 EDTA versus placebo, Outcome 24 Adverse event - hypocalcaemic symptoms. . . . 26 Analysis 1.25. Comparison 1 EDTA versus placebo, Outcome 25 Adverse event - fatigue...... 26 Analysis 1.26. Comparison 1 EDTA versus placebo, Outcome 26 Adverse event - faintness...... 27 Analysis 1.27. Comparison 1 EDTA versus placebo, Outcome 27 Adverse event - gastrointestinal symptoms. . . . 27 Analysis 1.28. Comparison 1 EDTA versus placebo, Outcome 28 Adverse event - serum-creatinine increase. . . . 28 Analysis 1.29. Comparison 1 EDTA versus placebo, Outcome 29 Adverse event - proteinuria...... 28 Analysis 1.30. Comparison 1 EDTA versus placebo, Outcome 30 Adverse event - phlebitis at infusion site. . . . . 29 Analysis 1.31. Comparison 1 EDTA versus placebo, Outcome 31 Adverse event - pain at infusion site...... 29 Analysis 1.32. Comparison 1 EDTA versus placebo, Outcome 32 Adverseevent-headache...... 30 Analysis 1.33. Comparison 1 EDTA versus placebo, Outcome 33 Adverse event - Raynaud’s phenomenon. . . . . 30

Chelation therapy for atherosclerotic cardiovascular disease (Review) i Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.34. Comparison 1 EDTA versus placebo, Outcome 34 Adverse event - metallic taste...... 31 Analysis 1.35. Comparison 1 EDTA versus placebo, Outcome 35 Adverse event - dermatitis...... 31 APPENDICES ...... 31 WHAT’SNEW...... 32 HISTORY...... 32 CONTRIBUTIONSOFAUTHORS ...... 32 DECLARATIONSOFINTEREST ...... 32 SOURCESOFSUPPORT ...... 32 INDEXTERMS ...... 33

Chelation therapy for atherosclerotic cardiovascular disease (Review) ii Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. [Intervention Review] Chelation therapy for atherosclerotic cardiovascular disease

Antonio L Dans2, Flordeliza N Tan3, Essie C Villarruz-Sulit1

1Section of Adult Medicine & Medical Research Unit, Department of Medicine, Ermita, Philippines. 2Section of Adult Medicine, College of Medicine, University of the Philippines, Ermita, Philippines. 3Manila, Philippines

Contact address: Essie C Villarruz-Sulit, Section of Adult Medicine & Medical Research Unit, Department of Medicine, 1st Floor, Philippine General Hospital, Taft Avenue, Ermita, Manila, Philippines. [email protected]. [email protected]. (Editorial group: Cochrane Peripheral Vascular Diseases Group.)

Cochrane Database of Systematic Reviews, Issue 3, 2009 (Status in this issue: Unchanged) Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. DOI: 10.1002/14651858.CD002785 This version ﬁrst published online: 21 October 2002 in Issue 4, 2002. Last assessed as up-to-date: 10 May 2005. (Help document - Dates and Statuses explained)

This record should be cited as: Dans AL, Tan FN, Villarruz-Sulit EC. Chelation therapy for atherosclerotic cardiovascular disease. Cochrane Database of Systematic Reviews 2002, Issue 4. Art. No.: CD002785. DOI: 10.1002/14651858.CD002785.

ABSTRACT Background Chelation therapy is being promoted and practiced all over the world as a form of alternative medicine in the treatment of atherosclerotic cardiovascular disease. It has been recommended as a safe, relatively inexpensive and non-surgical method of restoring blood flow in atherosclerotic vessels. At present the benefit of chelation therapy remains controversial at best. Objectives To assess the effects of ethylene diamine tetraacetic acid (EDTA) chelation therapy on clinical outcomes among people with atherosclerotic cardiovascular disease. Search strategy The Cochrane Peripheral Vascular Diseases Group Trials Register was searched, (last searched May 2005), the Cochrane Controlled Trials Register, (The Cochrane Library Issue 2, 2005), MEDLINE and EMBASE for published articles and other relevant articles. Studies were also requested through correspondence with known Filipino practitioners of the procedure. Selection criteria Studies were included if they were randomized controlled trials of EDTA chelation therapy versus placebo or no treatment in participants with atherosclerotic cardiovascular disease. Main outcome measures considered included either total or cause-specific mortality, nonfatal cardiovascular events, direct or indirect measurement of disease severity, subjective measures of improvement or adverse events. Data collection and analysis Two authors (MVV, FT) extracted data and assessed trial quality independently. Unresolved issues were considered by a third author (ALD). Discrepancies were discussed until a consensus was reached. Authors were contacted for additional information. Main results A total of five studies were included in the review. Mortality, non-fatal events, and cerebrovascular events were not reported in any of the studies. Four of the studies, with a total recruitment rate of 250 participants, showed no significant difference in the following outcomes: direct or indirect measurement of disease severity and subjective measures of improvement. One of the studies, which included only

Chelation therapy for atherosclerotic cardiovascular disease (Review) 1 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 10 participants, was interrupted prematurely, because of an apparent treatment effect. However, relevant data were not available in the report and have been requested from the authors.

Authors’ conclusions

At present, there is insufﬁcient evidence to decide on the effectiveness or ineffectiveness of chelation therapy in improving clinical outcomes of people with atherosclerotic cardiovascular disease. This decision must be preceded by conducting randomized controlled trials that would include endpoints that show the effects of chelation therapy on longevity and quality of life among people with atherosclerotic cardiovascular disease.

PLAIN LANGUAGE SUMMARY

Chelation therapy for atherosclerotic cardiovascular disease

Not enough evidence about the effects of chelation therapy to reduce blockages in the blood vessels of people with atherosclerotic cardiovascular (heart and circulation) disease. Atherosclerosis is caused by fatty deposits sticking to the inside of people’s arteries and restricting blood ﬂow. People with blocked arteries are more likely to have strokes and heart attacks, and can often only walk short distances before their legs begin to ache. Chelation therapy involves infusions into the bloodstream of substances believed to remove metals from the blood. This is promoted to people with atherosclerotic heart and circulation disease as a way of breaking down the blockages in their blood vessels. However, the review found there is not enough evidence from trials about the effects of this treatment.

BACKGROUND atherosclerosis has not been fully documented. In a book by Drs. Chelation therapy is being promoted and practiced as a form of Walker and Shah on chelation therapy (Walker 1997), the authors alternative medicine to treat atherosclerosis and relieve symptoms correctly noted that, “all possible mechanisms of action of chela- of cardiovascular disease (Hiatt 1997). It has been recommended tion therapy for producing the observed beneficial effects are still as a safe, relatively inexpensive and non-surgical method of restor- incompletely documented. And this incomplete understanding of ing blood flow in atherosclerotic vessels, thus preventing chronic why and how it works becomes a useful argument employed by symptoms such as angina pectoris and claudication, or acute symp- medical opponents of the method....There has, in fact, been no toms such as myocardial infarction or stroke. Several clinics have full-scale study of the technique.” been set up worldwide advocating its use in people with atherosclerotic cardiovascular disease. Thus, the authors decided to perform a systematic review of available information on this treatment in order to discuss implications Chelation therapy consists of a series of intravenous infusions con- for clinical practice and research. taining disodium ethylene diamine tetraacetic acid (EDTA) in combination with other substances. EDTA, a water soluble compound, has been found to be effective in chelating and removing OBJECTIVES toxic metals from the blood (Green 1993). It is capable of combining with polyvalent cations, such as calcium ions, to form a To determine whether EDTA chelation therapy is better than no soluble non-ionic complex that can be excreted (Wilder 1962). treatment or placebo in terms of improving clinical outcomes in Proponents of chelation believe that this mechanism, taking place people with atherosclerotic cardiovascular disease. at the arterial wall, can lead to regression of atherosclerotic plaques (Clarke 1960; Green 1993; Meltzer 1960). There have been case reports suggesting that EDTA chelation therapy in people with METHODS angina may lead to alleviation of symptoms (Clarke 1956; Meltzer 1960). On the other hand, a recent review on chelation therapy for peripheral arterial occlusive disease has shown that chelation Criteria for considering studies for this review therapy is not superior to placebo and is associated with consider- able risks (Ernst 1997). Types of studies At present, the benefit of chelation therapy in people with

Chelation therapy for atherosclerotic cardiovascular disease (Review) 2 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. All randomized controlled trials of chelation therapy compared to other published materials such as books and conference proceed- placebo or no treatment were included in this review. ings. The bibliographies of the studies thus retrieved were searched for further trials. There were no restrictions on language. Types of participants Trials on people with atherosclerotic cardiovascular disease - either cerebrovascular disease, coronary artery disease, or peripheral vas- Data collection and analysis cular disease. Most trials were on people with intermittent claudi- Selection of trials cation. All identified trials were checked by two authors (MVV, FT), to determine relevant articles for full text retrieval. Retrieved studies Types of interventions were then assessed for eligibility by the same two authors, according to the inclusion criteria specified. Issues that were unresolved Trials evaluating intravenous infusions containing EDTA com- in the assessment and data extraction process were passed on to pared to placebo infusions or no treatment. the third author (ALD). Discrepancies were discussed until a consensus was reached. Types of outcome measures Quality of trials The following outcome measures were considered: all cause mor- For each trial fulfilling the inclusion requirements, an assessment tality, coronary heart disease (CHD) deaths, and vascular deaths. of methodological quality was undertaken independently by two Non-fatal events including acute coronary syndromes such as my- authors (MVV, FT). A quality assessment instrument developed ocardial infarction and unstable angina pectoris as well as cere- by the Philippine Cardiovascular Research Group was used. This brovascular events such as stroke were also considered. Effects on instrument classifies study quality according to selection, detec- direct measurement of disease severity (e.g. digital subtraction an- tion, performance and exclusion biases. Specific features evaluated giograms for peripheral arterial disease), indirect tests of disease in this instrument were allocation concealment, balance in base- severity (e.g. ankle-brachial pressure index (ABPI)) and subjective line characteristics, blinding, drop-out rates, and analysis by in- measures of severity such as walking distance for claudicants were tention-to-treat. Issues that were unresolved in the assessment and also evaluated. Adverse events were also included. data extraction process were passed on to the third author (ALD). Discrepancies were discussed until a consensus was reached. Data extraction Search methods for identification of studies Data for the specified outcomes were extracted independently by two authors (MVV,FT). The information collected from each trial The Cochrane Peripheral Vascular Diseases Group searched included study design, participant characteristics, interventions their Specialised Register (last searched May 2005) and the compared and outcomes measured. Discrepancies were discussed Cochrane Central Register of Controlled Trials (CENTRAL) in until a consensus was reached. The Cochrane Library (last searched Issue 2, 2005) for publications Statistical analysis describing (or which might describe) randomized controlled trials Statistical pooling of outcomes was conducted. Heterogeneity be- (RCTs) of chelation therapy against placebo, or reference medi- tween trials was tested. Analysis was performed using the statis- cations, for atherosclerotic cardiovascular disease. See Appendix 1 tical guidelines set by the Cochrane Peripheral Vascular Disease for details of the search strategy used to search CENTRAL. Group. For dichotomous outcomes, relative risk (RR) was used as The Specialised Register is constructed from electronic searches a measure of effect. Continuous scales of measurements used were of MEDLINE (1966 to date), EMBASE (1980 to date), and analysed as weighted mean difference (WMD). CINAHL (1982 to date), and through handsearching relevant journals. The full list of journals that have been handsearched, as well as the search strategies used are described in the ’Search strategies for the identification of studies’ section within the editorial RESULTS information about the Cochrane PVD Group in The Cochrane Library, http:/ / www.mrw.interscience.wiley.com/ cochrane/ clabout/ Description of studies articles/ PVD/ frame.html Other terms were sought by visiting chelation clinics and deter- See: Characteristics of included studies; Characteristics of excluded mining the names of the manufactured agents used, particularly studies; Characteristics of ongoing studies. for chelation therapy on people with atherosclerotic cardiovascular A total of seven randomized controlled trials were retrieved ( disease. Studies were also requested through correspondence with Guldager 1992; Guldager (Dan) 1992; Guldager 1993; Guldager known Filipino “experts” in the procedure, and through reading of 1996; Olszewer 1990; Sloth-Nielsen 1991; Van Rij 1994). None

Chelation therapy for atherosclerotic cardiovascular disease (Review) 3 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. of the studies evaluated people with coronary or cerebrovascular Van Rij recruited 32 participants in the study (Van Rij 1994). disease. All seven studies recruited participants with peripheral Fifteen participants were randomized to chelation and 17 to the vascular disease, particularly those with intermittent claudication. control group. The major endpoints monitored were walking dis- A double publication was noted as the Guldager study in 1992 was tance (based on onset of pain or disabling claudication), subjective also published in Danish (Guldager (Dan) 1992). Two of the seven walking distance, or ABPI at rest, and on exertion. There was no studies (Guldager 1993; Guldager 1996), were actually substud- significant difference between the control and treatment groups ies of another (Guldager 1992). One of the substudies (Guldager for these major endpoints. Other endpoints included lifestyle mea- 1996), focused on mechanistic outcomes such as urinary metal sures (i.e. quality of life and physical well-being). Again, results excretion and magnesium retention and was eventually excluded. showed no significant difference. Thus, only five studies eventually underwent appraisal and data There were two studies authored by Guldager et al that re- extraction. cruited participants with intermittent claudication (Guldager The following outcomes were monitored: digital subtraction an- 1992; Guldager 1993). However, one of these was a substudy that giogram, walking distance based on onset of pain or disabling monitored the effect of EDTA on blood lipids (Guldager 1993). claudication, subjective walking distance, ankle brachial pressure In this study, treatment with EDTA did not alter total cholesterol, index (ABPI), lifestyle measures (i.e. quality of life and physical high density lipoprotein (HDL), low density lipoprotein (LDL) or well-being), blood lipid levels, pain-free maximal walking distance triglycerides. Guldager’s main trial of 153 participants monitored measured on treadmill, or walking and bicycle stress tests. pain-free and maximal walking distance measured on treadmill, as Mortality was not reported in any of the trials. Non-fatal events well as ABPI (Guldager 1992). The results failed to demonstrate such as myocardial infarction, unstable angina pectoris and cere- any significant effect. There was no significant difference as well, in brovascular events were not included as endpoints in any of the terms of subjective evaluation by participants. Adverse events were studies. reported but not subjected to statistical analysis. These potential All studies compared EDTA to placebo. The placebo was either side effects were as follows: stroke, hypocalcemic symptoms, fa- an isotonic solution or distilled water. tigue, faintness, gastrointestinal symptoms, serum creatinine increase, proteinuria, phlebitis at infusion site, pain at infusion site, headache, Raynaud’s phenomenon, metallic taste and dermatitis. Only one article reported improvement in the application of chela- Risk of bias in included studies tion therapy (Olszewer 1990). The study by Olszewer, et al. started The quality of the included studies ranged from A to C, based on with a randomized, double-blind method, but was eventually com- the Philippine Cardiovascular Research Group assessment scale. pleted as a single-blind study with no control group. Ten partici- Allocation concealment and double-blinding were stated or im- pants were recruited and initially received either EDTA or placebo. plied in all of the studies. However, in the study by Olszewer ( The study was arbitrarily stopped because, according to the in- Olszewer 1990), the code was broken before the study’s comple- vestigators, “after 10 treatments it was apparent that some partic- tion, thereby opening the study to performance, exclusion, and al- ipants were improving substantially but others were not ..... we location bias. Only the study by Van Rij (Van Rij 1994) described therefore, decided to break the code.” The investigators then dean intention-to-treat type of analysis at three months follow-up. cided to finish the study on a single-blind basis with no control group, convinced with the improvements obtained from EDTA treatment. Data from this study cannot be included in the statistical analysis as standard deviations were not reported in the article. Effects of interventions We asked the trial authors for more data but have not received any response to date. All the studies evaluated the use of EDTA on people with periph- Overall, there were no significant changes in ABPI, walking dis- eral arterial disease. The study by Sloth-Nielsen analyzed a group tance, or pain-free walking distance. Weighted mean difference of 30 participants (Sloth-Nielsen 1991). The outcome reported (WMD) for ABPI was 0.01 (95% confidence interval (CI) -0.03 focused on digital subtraction angiograms. The result showed no to 0.06) at the 20th infusion (Guldager 1992; Van Rij 1994), and significant difference between EDTA chelation and placebo. Other 0.02 (95% CI -0.03 to 0.06) at three months post-treatment ( outcomes were measured such as walking distance, ABPI and tran- Guldager 1992; Van Rij 1994). WMD for walking distance was - scutaneous tension of oxygen diffusion on the skin. Exact data for 37.93 (95% CI -90.32 to 14.45) at the 20th infusion (Guldager analysis could not be extracted from the article. Further informa- 1992; Van Rij 1994), and -31.46 (95% CI -87.63 to 24.71) at three tion was requested but none has been forthcoming. However, the months post-treatment (Guldager 1992; Van Rij 1994). Lastly, trial authors concluded that EDTA chelation therapy has no ben- WMD for pain-free walking distance was -16.4 (95% CI -31.99 eficial effect on objective parameters or on subjective evaluations to -0.81) at the 20th infusion (Guldager 1992; Van Rij 1994) and on participants with claudication in the lower extremities caused -7.73 (95% CI -22.59 to 7.13) at three months post-treatment. by arteriosclerosis.

Chelation therapy for atherosclerotic cardiovascular disease (Review) 4 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. The side effects of chelation noted included faintness, relative risk (Chappel 1994) may be indicative of a treatment effect, however, (RR) 11.44 (95% CI 1.51 to 86.45) and possibly gastrointestinal benefit is still questionable in the absence of a large randomized symptoms, RR 1.63 (95% CI 0.67 to 3.99), proteinuria RR 2.60 controlled trial. (95% CI 0.85 to 7.93), and hypocalcemia, RR 3.12 (95% CI 0.65 to 14.98) (Guldager 1992). Procedure-related side effects were also noted including phlebitis and pain at the infusion site (Guldager AUTHORS’ CONCLUSIONS 1992). Implications for practice At the present time, there is insufficient evidence to decide on the effectiveness or ineffectiveness of chelation therapy in improving DISCUSSION clinical outcomes among people with atherosclerotic cardiovascular disease. Wider acceptance of this treatment in clinical practice Treatment with EDTA for metal poisoning has been established must be preceded by conducting more randomized clinical trials, for decades (Guldager 1996). However, since the 1950s chelation especially among people at risk for coronary or cerebrovascular therapy has been suggested for the treatment of coronary disease disease. and other vascular diseases (Clarke 1955; Clarke 1956). Based on the premise that since metastatic calcium deposits can be removed Implications for research by this technique, the calcium involved in atheromata could be evacuated in a similar way. This mechanism would eventually To date, there are no trials on the effectiveness of chelation therapy lead to an improvement in outcome of participants with coronary on coronary and cerebrovascular disease. Trials have centered on disease. Initial claims on the effect of EDTA in lowering serum peripheral vascular disease, more specifically treatment of inter- calcium levels were based on animal experiments (Clarke 1956). mittent claudication. Therefore, it is important that randomized However, it was the result of active administration of EDTA to trials be conducted in people for whom the treatment is intended people at that time that provided encouragement for the use of and supposed to benefit, i.e. people with coronary and cerebrovas- this form of therapy. The continued use and promotion of chela- cular disease. It is also important that future trials using chelation tion therapy, however, is hampered by the limited availability of therapy include endpoints that show its effects on longevity and randomized controlled trials that have shown benefit. quality of life rather than monitoring mechanistic outcomes.

The studies published to date have failed to demonstrate significant benefit with the use of EDTA chelation therapy in people with peripheral vascular disease. A single study showed benefit, ACKNOWLEDGEMENTS but conclusions were weak because of serious methodological flaws seen upon review (Olszewer 1990). Although a meta-analysis of The authors wish to thank their family, friends and colleagues in unpublished data reported a high correlation between improve- the profession for all the support. We are also grateful for the en- ment in cardiovascular function and treatment with EDTA, a ran- couragement given by the following people: Prof. Shah Ebrahim, domized controlled method was not used by the various investiga- Theresa Moore and Katherine Wornell of the Cochrane Heart tors in their recruitment of participants (Chappel 1994). There- Group. Our deepest appreciation goes to the Cochrane Periph- fore, contrary to circulating claims, no published randomized con- eral Vascular Disease Group, especially to Dr. Elizabeth Royle and trolled trials have evaluated the use of EDTA in participants with Heather Maxwell for the technical support, encouraging words coronary or cerebrovascular disease. The meta-analysis by Chappel and reminders.

REFERENCES

References to studies included in this review dication. A double-blind randomized trial [EDTA versus placebobe- handling af claudicatio intermittens]. Ugeskrift for Laeger 1992;154 (23):1618–21. Guldager 1992 {published data only} Guldager B, Jelnes R, Jorgensen SJ, Nielsen JS, Klaerke A, Mogensen K, et al.EDTA treatment of intermittent claudication - a double- Guldager 1993 {published data only} blind, placebo-controlled study. Journal of Internal Medicine 1992; Guldager B, Faergeman O, Jorgensen SJ, Nexo E, Jelnes R. Dis- 231(3):261–7. odium-ethylene diamine tetraacetic acid (EDTA) has no effect on Guldager B, Jelnes R, Jorgensen SJ, Sloth-Nielsen J, Klaerke A, Mo- blood lipids in atherosclerotic patients. A randomized, placebo-con- gensen K, et al.EDTA versus placebo treatment in intermittent clau- trolled study. Danish Medical Bulletin 1993;40(5):625–7.

Chelation therapy for atherosclerotic cardiovascular disease (Review) 5 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Olszewer 1990 {published data only} Clarke 1955 Olszewer E, Sabbag FC, Carter JP. A pilot double-blind study of Clarke NE, Clarke CN, Mosher RE. The “in vivo” dissolution of sodium magnesium EDTA in peripheral vascular disease. Journal of metastatic calcium. An approachtoatherosclerosis. AmericanJournal the National Medical Association 1990;82(3):173–7. of Medical Sciences 1955;229(2):142–9. Sloth-Nielsen 1991 {published data only} Clarke 1956 Sloth-Nielsen J, Guldager B, Mouritzen C, Lund EB, Egebland M, Clarke NE, Clarke CN, Mosher RE. Treatment of angina pectoris Norregaard O, et al.Arteriographic ﬁndings in EDTA chelation ther- with disodium ethylene diamine tetraacetic acid. American Journal apy on peripheral arteriosclerosis. American Journal of Surgery 1991; of Medical Sciences 1956;232(6):654–66. 162(2):122–5. Clarke 1960 Van Rij 1994 {published data only} Clarke N. Atherosclerosis, occlusive vascular disease and EDTA. The Van Rij AM, Solomon C, Packer SGK, Hopkins WG. Chelation American Journal of Cardiology 1960;6(2):1–3. therapy for intermittent claudication: A double-blind, randomized, Ernst 1997 controlled trial. Circulation 1994;90(3):1194–9. Ernst E. Chelation therapy for peripheral arterial occlusive disease: a systematic review. Circulation 1997;96(3):1031–3. References to studies excluded from this review Green 1993 Green S. Chelation Therapy: unproven claims and unsound theories. Anderson 2003 {published data only} Nutrition Forum 1993. Anderson TJ, Hubacek J, Wyse DG, Knudtson ML. Effect of chela- Guldager (Dan) 1992 tion therapy on endothelial function in patients with coronary artery Guldager B, Jelnes R, Jorgensen SJ, Sloth-Nielsen J, Klaerke A, Mo- disease: PATCH substudy. Journal of the American College of Cardi- gensen K, et al.EDTA versus placebo treatment in intermittent clau- ology 2003;41(3):420–5. dication. A double-blind randomized trial [EDTA versus placebobe- Guldager 1996 {published data only} handling af claudicatio intermittens]. Ugeskrift for Laeger 1992;154 Guldager B, Jorgensen PJ, Grandjean P. Metal excretion and mag- (23):1618–21. nesium retention in patients with intermittent claudication treated Hiatt 1997 with intravenous disodium EDTA. Clinical Chemistry 1996;42(12): Hiatt WR. Current and future drug therapies for claudication. Vas- 1938–42. cular Medicine 1997;2(3):257–62. References to ongoing studies Meltzer 1960 Meltzer LE, Ural ME, Kitchell JR. The treatment of coronary artery disease with disodium EDTA. Metal-Binding in Medicine, edited by TACT 2003 {published data only} Seven MJ. Philadelphia: Lippincott, 1960. Trial to Assess Chelation Therapy (TACT). http: //www.clinicaltrials.gov/ct/show/NCT00044213?order=1. Walker 1997 Walker M, Shah H. Everything you should know about chelation ther- Additional references apy. New Canaan, Connecticut: Keats Publishing, 1997. Wilder 1962 Chappel 1994 Wilder LW, De Jode LR, Milstein SW, Howard JM. Mobilization Chappel LT, Stahl JP, Evans R. EDTA chelation treatment for vas- of atherosclerotic plaque calcium with EDTA utilizing the isolation- cular disease: a meta-analysis using unpublished data. Journal of Ad- perfusion principle. Surgery 1962;52(5):793–5. vancement in Medicine 1994;7(3):131–42. ∗ Indicates the major publication for the study

Characteristics of included studies [ordered by study ID]

Guldager 1992

Methods Study design: Randomized, double-blind trial. Method of randomization: patients randomized in blocks of 10. Exclusions post-randomization: 6 (5 EDTA, 1 placebo). Losses to follow up: 4 (2 deteriorations leading to vascular surgery, 1 death, 1 patient started chelation therapy at a private clinic).

Participants Country: Denmark. Setting: Outpatient clinic. No. of patients: 159 (80 EDTA, 79 placebo). Age: 64 ± 7 years EDTA, 66 ± 9 years placebo. Gender: 48M : 32F EDTA, 55M : 24F placebo. Inclusion criteria: stable intermittent claudication for at least 12 months; pain-free walking distance range of 50 to 200 m measured on treadmill at a speed of 3.6 km/hr with 10 degree inclination; ABPI of worse leg < 0.8. Exclusion criteria: Vascular surgery within last 12 months, ischaemic rest pain or gangrene, moderate or severe venous insufficiency, renal insufficiency, diabetes mellitus, thyroid or parathyroid disorders, hepatic dysfunction, significant cardio-pulmonary failure, eg acute myocardial infarction within last 12 months, tuberculosis, pregnancy, other conditions which could limit the patient’s walking distance or reliable interpretation of the study, patients receiving anticoagulants, nitroglycerine or lithium, EDTA chelation therapy within last 24 months.

Interventions Treatment: EDTA 3 g + NaCl 8.4 g in 1 litre normal saline solution x 20 infusions. Control: 1 litre normal saline solution x 20 infusions. Duration of treatment: 5-9 weeks. Follow up: 3 and 6 months.

Outcomes 1. Subjective evaluation. 2. Pain-free and maximal walking distances. 3. ABPI (6 months duration of observation).

Notes

Risk of bias

Item Authors’ judgement Description

Allocation concealment? Yes A - Adequate

Methods Study design: Randomized, double-blind trial or parallel group design. Method of randomization: not stated but presumably as per Guldager 1992, patients randomized in blocks of 10. Exclusions post-randomization: Not stated. Losses to follow up: Not stated.

Participants Country: Denmark. Setting: Outpatient clinic. No. of patients: 29 (14 EDTA, 15 placebo). Age: over 40 years. Gender: Not stated. Inclusion criteria: stable intermittent claudication for 1 year; systolic ABPI < 0.8; pain-free walking distance 50 to 200 m. Exclusion criteria: Diabetes mellitus, rest pain, ulcers or gangrene, kidney or liver disease, alcoholism.

Interventions Treatment: EDTA 3 g in 1 litre isotonic solution x 20 infusions. Control: 1 litre isotonic saline x 20 infusions. Patients also received daily vitamins, minerals and trace elements. Duration of treatment: 5-9 weeks. Follow up: Not stated.

Outcomes Plasma concentration of cholesterol, LDL, HDL, or triglycerides, (5-9 weeks duration of observation).

Notes Substudy of Guldager 1992.

Risk of bias

Item Authors’ judgement Description

Allocation concealment? Yes A - Adequate

Olszewer 1990

Methods Study design: Randomized, double-blind for 10 treatments, then completed in an open single treatment fashion. Method of randomization: Ampules numbered 1 to 10, patients randomly and equally divided; identiﬁ- cation of contents of each vial in a sealed envelope held by the manufacturer of the ampules. Exclusions post-randomization: Not stated. Losses to follow up: Not stated.

Participants Country: Brazil/USA. Setting: Outpatient clinic. No. of patients: 10. Age: Mean 47 years (range 41 to 53). Gender: Male. Inclusion criteria: peripheral vascular disease (Fontaine Stage II) - intermittent claudication; walking test - claudication between 100 and 300 m; master exercise test - claudication with < 40 steps; bicycle stress test - claudication before 3 minutes at 50 km/hr. Exclusion criteria: Pain at rest or at night, gangrene.

Interventions Treatment: EDTA 10 ml (1.5 g). Control: Distilled water 10 ml. Duration of treatment: 20 infusions (no time length speciﬁed). Follow up: Not stated.

Outcomes 1. Walking distance measured by Walking Test. 2. Number of steps measured by Master Exercise Test. 3. Cycling time at 25 km/hr by Bicycle Test, (observations made after 10 infusions and after 20 infusions).

Notes Adequate allocation concealment initially, but the code was broken before end of study.

Risk of bias

Item Authors’ judgement Description

Allocation concealment? Yes A - Adequate

Sloth-Nielsen 1991

Methods Study design: Randomized, double-blind trial. Method of randomization: Blocks of 10 according to a randomization code. Exclusions post-randomization: None. Losses to follow up: None.

Participants Country: Denmark. Setting: Outpatient clinic. No. of patients: 30. Age: > 40 years old. Gender: Not stated. Inclusion criteria: (+) stable intermittent claudication; walking range of 50 to 200 m; ABPI < 0.8. Exclusion criteria: vascular surgery within the last 12 months, ischaemic rest pain or gangrene, moderate or severe venous insufﬁciency, renal insufﬁciency, diabetes mellitus, thyroid or parathyroid disorders,

hepatic dysfunction, signiﬁcant cardio-pulmonary failure (eg. acute myocardial infarction within last year), coexistent carcinomas, tuberculosis, pregnancy, other conditions that could limit the patient’s walking distance or the full understanding of the study, patients receiving anticoagulants, nitroglycerin or lithium, EDTA chelation therapy within the last 24 months.

Interventions Treatment: EDTA 3 g + NaCl 8.4 g in sterile water; 1 litre isotonic solution x 20 infusions. Control: 1 litre isotonic solution x 20 infusions. Patients also received oral viatmins, minerals and trace elements. Duration of treatment: 6-10 weeks. Follow up: 3 months.

Outcomes Digital subtraction angiograms, (6 to 10 weeks duration of observation).

Notes Substudy of a larger trial not speciﬁed within the article.

Risk of bias

Item Authors’ judgement Description

Allocation concealment? Yes A - Adequate

Van Rij 1994

Methods Study design: Randomized, double-blind trial. Method of randomization: blocks of 10. Exclusions post-randomization: None up to 3 month assessment. Losses to follow up: None up to 3 month assessment.

Participants Country: New Zealand. Setting: Outpatient clinic. No. of patients: 32 (15 treatment, 17 control). Age: 67.7 ± 7.0 years EDTA, 66.9 ± 6.7 years control. Gender: 13M : 2F EDTA, 15M : 2F control. Inclusion criteria: Intermittent claudication < 20% variation in measured walking distance over 3 consecutive assessments performed on different days, older than 45 years. Exclusion criteria: Other debilitating disease that affected walking, signiﬁcant renal disease, diabetes mellitus.

Interventions Treatment: EDTA 3 g + MgCl 0.76 g + NaHCO3 0.84 g in 500 ml normal saline x 20 infusions. Control: 500 ml normal saline x 20 infusions.

Both groups’ infusions contained Parentrovite (thiamine hydrochloride 250 mg, riboﬂavine phosphate 5.5 mg, pyridoxine hydrochloride 50 mg, ascorbic acid 500 mg, nicotinamide 160 mg, sodium pantothenate 5 mg, glucose anhydrous 1000 mg, sodium metabisulphite 4 mg). Both groups also received oral daily vitamin supplements. Duration of treatment: 10 weeks. Follow up: after 10 infusions, 3, 6 and 12 months (6 and 12 month assessments still in progress at time of article). Outcomes 1. Measured walking distance as total distance the patient was able to walk at 4 km/hr on a treadmill at 10% gradient to onset of pain or before stopping because of claudication. 2. Subjective walking distance as distance the patients considered able to walk before stopping because of claudication. 3. Ankle/brachial indices at rest and immediately after TET, (12 weeks duration of observation).

Notes

Risk of bias

Item Authors’ judgement Description

Allocation concealment? Yes A - Adequate

ABPI Ankle-brachial pressure index EDTA Ethylene diamine tetra-acetic acid HDL High density lipoprotein LDL Low density lipoprotein MgCl Magnesium chloride NaCl Sodium chloride NaHCO3 Sodium hydrogen carbonate TET Treadmill exercise test

Characteristics of excluded studies [ordered by study ID]

Anderson 2003 Outcome measured was ﬂow-mediated vasodilation.

Guldager 1996 Reported only on metal excretion and magnesium retention.

Chelation therapy for atherosclerotic cardiovascular disease (Review) 11 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Characteristics of ongoing studies [ordered by study ID]

TACT 2003

Trial name or title Trial to Assess Chelation Therapy (TACT).

Methods

Participants Patients with coronary artery disease.

Interventions 40 infusions of standard chelation solution versus 40 infusions of placebo in a 2x2 factorial design.

Outcomes Composite outcome of all cause mortality, myocardial infarction (MI), stroke, hospitalization for angina and hospitalization for congestive heart failure. Secondary endpoints will be: a) combined outcome of cardiac death or nonfatal MI or nonfatal stroke; b) individual components of the primary outcome; c) coronary revascularisation; d) safety of interventions including renal, hepatic and hematological function; e) health related quality of life; f) cost and cost-effectiveness; g) brachial artery ﬂow-mediated endothelial function; h) plasma markers of oxidative stress and anti-oxidant protection; i) plasma markers of endothelial activation and inﬂammation.

Starting date March 2003.

Contact information NCCAM Clearinghouse +1 888 644 6226 or http://www.clinicaltrials.gov/ct/show/NCT00044213?order=1

Notes

Comparison 1. EDTA versus placebo

No. of No. of Outcome or subgroup title studies participants Statistical method Effect size

1 Ankle-brachial pressure index at 2 185 Mean Difference (IV, Fixed, 95% CI) 0.01 [-0.03, 0.06] 20th infusion 2 Ankle-brachial pressure index at 2 181 Mean Difference (IV, Fixed, 95% CI) 0.02 [-0.03, 0.06] 3 months post-treatment 3 Ankle-brachial pressure index at 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected 6 months post-treatment 4 Walking distance (m) at 10th 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected infusion 5 Walking distance (m) at 20th 2 167 Mean Difference (IV, Fixed, 95% CI) -37.93 [-90.32, infusion 14.45] 6 Walking distance (m) at 3 2 165 Mean Difference (IV, Fixed, 95% CI) -31.46 [-87.63, months post-treatment 24.71] 7 Walking distance (m) at 6 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected months post-treatment 8 Pain-free walking distance (m) at 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected 10th infusion 9 Pain-free walking distance (m) at 2 167 Mean Difference (IV, Fixed, 95% CI) -16.40 [-31.99, - 20th infusion 0.81] 10 Pain-free walking distance (m) 2 165 Mean Difference (IV, Fixed, 95% CI) -7.73 [-22.59, 7.13] at 3 months post-treatment 11 Pain-free walking distance (m) 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected at 6 months post-treatment 12 Subjective walking distance (m) 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected at 20th infusion 13 Subjective walking distance (m) 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected at 3 months post-treatment 14 Subjective evaluation - 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected aggravation 15 Subjective evaluation - 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected unchanged 16 Subjective evaluation - mild 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected improvement 17 Subjective evaluation - 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected moderate improvement 18 Subjective evaluation - great 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected improvement 19 Plasma low density lipoprotein 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected (LDL)-cholesterol levels (mmol/litre) 20 Plasma triglyceride levels 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected (mmol/litre)

Chelation therapy for atherosclerotic cardiovascular disease (Review) 13 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 21 Plasma high density lipoprotein 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected (HDL)-cholesterol levels (mmol/litre) 22 Plasma cholesterol levels 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected (mmol/litre) 23 Adverse event - stroke 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected 24 Adverse event - hypocalcaemic 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected symptoms 25 Adverse event - fatigue 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected 26 Adverse event - faintness 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected 27 Adverse event - gastrointestinal 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected symptoms 28 Adverse event - serum- 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected creatinine increase 29 Adverse event - proteinuria 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected 30 Adverse event - phlebitis at 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected infusion site 31 Adverse event - pain at infusion 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected site 32 Adverse event - headache 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected 33 Adverse event - Raynaud’s 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected phenomenon 34 Adverse event - metallic taste 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected 35 Adverse event - dermatitis 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected

Analysis 1.1. Comparison 1 EDTA versus placebo, Outcome 1 Ankle-brachial pressure index at 20th infusion.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 1 Ankle-brachial pressure index at 20th infusion

Studyorsubgroup Treatment Control MeanDifference Weight Mean Difference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 75 0.52 (0.14) 78 0.51 (0.13) 95.4 % 0.01 [ -0.03, 0.05 ]

Van Rij 1994 15 0.7 (0.36) 17 0.6 (0.15) 4.6 % 0.10 [ -0.10, 0.30 ] Total (95% CI) 90 95 100.0 % 0.01 [ -0.03, 0.06 ] Heterogeneity: Chi2 = 0.78, df = 1 (P = 0.38); I2 =0.0% Test for overall effect: Z = 0.66 (P = 0.51)

-20 -10 0 10 20 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 14 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.2. Comparison 1 EDTA versus placebo, Outcome 2 Ankle-brachial pressure index at 3 months post-treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 2 Ankle-brachial pressure index at 3 months post-treatment

Studyorsubgroup Treatment Control MeanDifference Weight Mean Difference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 73 0.54 (0.14) 76 0.53 (0.14) 82.6 % 0.01 [ -0.03, 0.05 ]

Van Rij 1994 15 0.62 (0.15) 17 0.58 (0.13) 17.4 % 0.04 [ -0.06, 0.14 ] Total (95% CI) 88 93 100.0 % 0.02 [ -0.03, 0.06 ] Heterogeneity: Chi2 = 0.30, df = 1 (P = 0.59); I2 =0.0% Test for overall effect: Z = 0.73 (P = 0.47)

-10 -5 0 5 10 Favours treatment Favours control

Analysis 1.3. Comparison 1 EDTA versus placebo, Outcome 3 Ankle-brachial pressure index at 6 months post-treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 3 Ankle-brachial pressure index at 6 months post-treatment

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 57 0.56 (0.13) 66 0.53 (0.14) 0.03 [ -0.02, 0.08 ]

-10 -5 0 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 15 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.4. Comparison 1 EDTA versus placebo, Outcome 4 Walking distance (m) at 10th infusion.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 4 Walking distance (m) at 10th infusion

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 70 164 (79) 73 192 (222) -28.00 [ -82.18, 26.18 ]

-1000 -500 0 500 1000 Favours control Favours treatment

Analysis 1.5. Comparison 1 EDTA versus placebo, Outcome 5 Walking distance (m) at 20th infusion.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 5 Walking distance (m) at 20th infusion

Studyorsubgroup Treatment Control MeanDifference Weight Mean Difference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 68 159 (99) 67 206 (239) 71.7 % -47.00 [ -108.88, 14.88 ]

Van Rij 1994 15 208 (135) 17 223 (149) 28.3 % -15.00 [ -113.41, 83.41 ] Total (95% CI) 83 84 100.0 % -37.93 [ -90.32, 14.45 ] Heterogeneity: Chi2 = 0.29, df = 1 (P = 0.59); I2 =0.0% Test for overall effect: Z = 1.42 (P = 0.16)

-1000 -500 0 500 1000 Favours control Favours treatment

Chelation therapy for atherosclerotic cardiovascular disease (Review) 16 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.6. Comparison 1 EDTA versus placebo, Outcome 6 Walking distance (m) at 3 months post- treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 6 Walking distance (m) at 3 months post-treatment

Studyorsubgroup Treatment Control MeanDifference Weight Mean Difference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 66 162 (101) 67 204 (248) 76.6 % -42.00 [ -106.19, 22.19 ]

Van Rij 1994 15 233 (138) 17 230 (195) 23.4 % 3.00 [ -113.06, 119.06 ] Total (95% CI) 81 84 100.0 % -31.46 [ -87.63, 24.71 ] Heterogeneity: Chi2 = 0.44, df = 1 (P = 0.51); I2 =0.0% Test for overall effect: Z = 1.10 (P = 0.27)

-1000 -500 0 500 1000 Favours control Favours treatment

Analysis 1.7. Comparison 1 EDTA versus placebo, Outcome 7 Walking distance (m) at 6 months post- treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 7 Walking distance (m) at 6 months post-treatment

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 51 180 (150) 56 194 (127) -14.00 [ -66.93, 38.93 ]

-1000 -500 0 500 1000 Favours control Favours treatment

Chelation therapy for atherosclerotic cardiovascular disease (Review) 17 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.8. Comparison 1 EDTA versus placebo, Outcome 8 Pain-free walking distance (m) at 10th infusion.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 8 Pain-free walking distance (m) at 10th infusion

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 70 97 (41) 73 101 (53) -4.00 [ -19.49, 11.49 ]

-100 -50 0 50 100 Favours control Favours treatment

Analysis 1.9. Comparison 1 EDTA versus placebo, Outcome 9 Pain-free walking distance (m) at 20th infusion.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 9 Pain-free walking distance (m) at 20th infusion

Studyorsubgroup Treatment Control MeanDifference Weight Mean Difference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 68 93 (40) 67 109 (56) 90.0 % -16.00 [ -32.44, 0.44 ]

Van Rij 1994 15 101 (50) 17 121 (89) 10.0 % -20.00 [ -69.30, 29.30 ] Total (95% CI) 83 84 100.0 % -16.40 [ -31.99, -0.81 ] Heterogeneity: Chi2 = 0.02, df = 1 (P = 0.88); I2 =0.0% Test for overall effect: Z = 2.06 (P = 0.039)

-100 -50 0 50 100 Favours control Favours treatment

Chelation therapy for atherosclerotic cardiovascular disease (Review) 18 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.10. Comparison 1 EDTA versus placebo, Outcome 10 Pain-free walking distance (m) at 3 months post-treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 10 Pain-free walking distance (m) at 3 months post-treatment

Studyorsubgroup Treatment Control MeanDifference Weight Mean Difference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 66 95 (48) 67 102 (42) 93.9 % -7.00 [ -22.34, 8.34 ]

Van Rij 1994 15 104 (62) 17 123 (108) 6.1 % -19.00 [ -79.17, 41.17 ] Total (95% CI) 81 84 100.0 % -7.73 [ -22.59, 7.13 ] Heterogeneity: Chi2 = 0.14, df = 1 (P = 0.70); I2 =0.0% Test for overall effect: Z = 1.02 (P = 0.31)

-100 -50 0 50 100 Favours control Favours treatment

Analysis 1.11. Comparison 1 EDTA versus placebo, Outcome 11 Pain-free walking distance (m) at 6 months post-treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 11 Pain-free walking distance (m) at 6 months post-treatment

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1992 51 97 (47) 56 119 (93) -22.00 [ -49.56, 5.56 ]

-100 -50 0 50 100 Favours control Favours treatment

Chelation therapy for atherosclerotic cardiovascular disease (Review) 19 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.12. Comparison 1 EDTA versus placebo, Outcome 12 Subjective walking distance (m) at 20th infusion.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 12 Subjective walking distance (m) at 20th infusion

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Van Rij 1994 15 413 (775) 17 327 (461) 86.00 [ -363.27, 535.27 ]

-1000 -500 0 500 1000 Favours control Favours treatment

Analysis 1.13. Comparison 1 EDTA versus placebo, Outcome 13 Subjective walking distance (m) at 3 months post-treatment.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 13 Subjective walking distance (m) at 3 months post-treatment

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Van Rij 1994 15 448 (556) 17 381 (473) 67.00 [ -293.17, 427.17 ]

-1000 -500 0 500 1000 Favours control Favours treatment

Chelation therapy for atherosclerotic cardiovascular disease (Review) 20 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.14. Comparison 1 EDTA versus placebo, Outcome 14 Subjective evaluation - aggravation.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 14 Subjective evaluation - aggravation

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 1/75 0/78 3.12 [ 0.13, 75.37 ]

0.001 0.01 0.1 1 10 100 1000 Favours treatment Favours control

Analysis 1.15. Comparison 1 EDTA versus placebo, Outcome 15 Subjective evaluation - unchanged.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 15 Subjective evaluation - unchanged

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 38/75 39/78 1.01 [ 0.74, 1.39 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 21 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.16. Comparison 1 EDTA versus placebo, Outcome 16 Subjective evaluation - mild improvement.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 16 Subjective evaluation - mild improvement

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 19/75 23/78 0.86 [ 0.51, 1.44 ]

0.1 0.2 0.5 1 2 5 10 Favours control Favours treatment

Analysis 1.17. Comparison 1 EDTA versus placebo, Outcome 17 Subjective evaluation - moderate improvement.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 17 Subjective evaluation - moderate improvement

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 13/75 9/78 1.50 [ 0.68, 3.31 ]

0.1 0.2 0.5 1 2 5 10 Favours control Favours treatment

Chelation therapy for atherosclerotic cardiovascular disease (Review) 22 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.18. Comparison 1 EDTA versus placebo, Outcome 18 Subjective evaluation - great improvement.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 18 Subjective evaluation - great improvement

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 4/75 7/78 0.59 [ 0.18, 1.95 ]

0.1 0.2 0.5 1 2 5 10 Favours control Favours treatment

Analysis 1.19. Comparison 1 EDTA versus placebo, Outcome 19 Plasma low density lipoprotein (LDL)- cholesterol levels (mmol/litre).

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 19 Plasma low density lipoprotein (LDL)-cholesterol levels (mmol/litre)

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1993 14 4.37 (0.21) 15 4.85 (0.26) -0.48 [ -0.65, -0.31 ]

-10 -5 0 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 23 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.20. Comparison 1 EDTA versus placebo, Outcome 20 Plasma triglyceride levels (mmol/litre).

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 20 Plasma triglyceride levels (mmol/litre)

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1993 14 1.96 (0.17) 15 1.51 (0.18) 0.45 [ 0.32, 0.58 ]

-10 -5 0 5 10 Favours treatment Favours control

Analysis 1.21. Comparison 1 EDTA versus placebo, Outcome 21 Plasma high density lipoprotein (HDL)- cholesterol levels (mmol/litre).

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 21 Plasma high density lipoprotein (HDL)-cholesterol levels (mmol/litre)

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1993 14 1.03 (0.07) 15 1.43 (0.12) -0.40 [ -0.47, -0.33 ]

-10 -5 0 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 24 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.22. Comparison 1 EDTA versus placebo, Outcome 22 Plasma cholesterol levels (mmol/litre).

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 22 Plasma cholesterol levels (mmol/litre)

Studyorsubgroup Treatment Control MeanDifference MeanDifference N Mean(SD) N Mean(SD) IV,Fixed,95% CI IV,Fixed,95% CI

Guldager 1993 14 6.29 (0.24) 15 6.97 (0.25) -0.68 [ -0.86, -0.50 ]

-10 -5 0 5 10 Favours treatment Favours control

Analysis 1.23. Comparison 1 EDTA versus placebo, Outcome 23 Adverse event - stroke.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 23 Adverse event - stroke

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 1/75 0/78 3.12 [ 0.13, 75.37 ]

0.01 0.1 1 10 100 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 25 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.24. Comparison 1 EDTA versus placebo, Outcome 24 Adverse event - hypocalcaemic symptoms.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 24 Adverse event - hypocalcaemic symptoms

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 6/75 2/78 3.12 [ 0.65, 14.98 ]

0.01 0.1 1 10 100 Favours treatment Favours control

Analysis 1.25. Comparison 1 EDTA versus placebo, Outcome 25 Adverse event - fatigue.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 25 Adverse event - fatigue

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 12/75 11/78 1.13 [ 0.53, 2.41 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 26 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.26. Comparison 1 EDTA versus placebo, Outcome 26 Adverse event - faintness.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 26 Adverse event - faintness

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 11/75 1/78 11.44 [ 1.51, 86.45 ]

0.01 0.1 1 10 100 Favours treatment Favours control

Analysis 1.27. Comparison 1 EDTA versus placebo, Outcome 27 Adverse event - gastrointestinal symptoms.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 27 Adverse event - gastrointestinal symptoms

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 11/75 7/78 1.63 [ 0.67, 3.99 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 27 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.28. Comparison 1 EDTA versus placebo, Outcome 28 Adverse event - serum-creatinine increase.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 28 Adverse event - serum-creatinine increase

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 7/75 9/78 0.81 [ 0.32, 2.06 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Analysis 1.29. Comparison 1 EDTA versus placebo, Outcome 29 Adverse event - proteinuria.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 29 Adverse event - proteinuria

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 10/75 4/78 2.60 [ 0.85, 7.93 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 28 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.30. Comparison 1 EDTA versus placebo, Outcome 30 Adverse event - phlebitis at infusion site.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 30 Adverse event - phlebitis at infusion site

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 35/75 28/78 1.30 [ 0.89, 1.91 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Analysis 1.31. Comparison 1 EDTA versus placebo, Outcome 31 Adverse event - pain at infusion site.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 31 Adverse event - pain at infusion site

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 9/75 5/78 1.87 [ 0.66, 5.33 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 29 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.32. Comparison 1 EDTA versus placebo, Outcome 32 Adverse event - headache.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 32 Adverse event - headache

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 7/75 7/78 1.04 [ 0.38, 2.82 ]

0.1 0.2 0.5 1 2 5 10 Favours treatment Favours control

Analysis 1.33. Comparison 1 EDTA versus placebo, Outcome 33 Adverse event - Raynaud’s phenomenon.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 33 Adverse event - Raynaud’s phenomenon

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 1/75 0/78 3.12 [ 0.13, 75.37 ]

0.01 0.1 1 10 100 Favours treatment Favours control

Chelation therapy for atherosclerotic cardiovascular disease (Review) 30 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. Analysis 1.34. Comparison 1 EDTA versus placebo, Outcome 34 Adverse event - metallic taste.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 34 Adverse event - metallic taste

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 1/75 0/78 3.12 [ 0.13, 75.37 ]

0.01 0.1 1 10 100 Favours treatment Favours control

Analysis 1.35. Comparison 1 EDTA versus placebo, Outcome 35 Adverse event - dermatitis.

Review: Chelation therapy for atherosclerotic cardiovascular disease

Comparison: 1 EDTA versus placebo

Outcome: 35 Adverse event - dermatitis

Studyorsubgroup Treatment Control RiskRatio RiskRatio n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI Guldager 1992 1/75 0/78 3.12 [ 0.13, 75.37 ]

0.01 0.1 1 10 100 Favours treatment Favours control

APPENDICES

Appendix 1. Search strategy for CENTRAL 1) CARDIOVASCULAR DISEASES MeSH explode tree 1 2) (CARDIOVASCULAR next DISEASE*) 3) (PERIPHERAL next VASCULAR next DISEASE*) 4) #1 or #2 or #3 5) CHELATION THERAPY MeSH single term 6) CHELATING AGENTS MeSH explode tree 1 7) CHELATION* 8) ETHYLENEDIAMINES MeSH single term 9) (ETHYLENE next DIAMINE next TETRAACETIC next ACID*) 10) (ETHYLENE next DIAMINE next TETRA next ACETIC next ACID*)

Chelation therapy for atherosclerotic cardiovascular disease (Review) 31 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. 11) (DISODIUM next ETHYLENE next DIAMINE next TETRAACETIC next ACID) 12) EDTA 13) #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 14) (#4 and #13)

WHAT’S NEW Last assessed as up-to-date: 10 May 2005.

29 May 2008 Amended Converted to new review format.

HISTORY Protocol ﬁrst published: Issue 4, 2000 Review ﬁrst published: Issue 4, 2002

30 October 2006 Amended Minor edit

25 May 2005 New search has been performed Review updated with the addition of one ongoing study and minor style guide changes

CONTRIBUTIONSOFAUTHORS Villarruz MVC - primary author, contact reviewer, appraised studies and extracted data. Dans AL - author, reviewer, resolved discrepancies. Tan F - author, reviewer, appraised studies and extracted data.

DECLARATIONSOFINTEREST None known.

Internal sources

• No sources of support supplied

External sources

• The Chief Scientist Ofﬁce, Scottish Executive Health Directorates, The Scottish Government, UK.

INDEX TERMS

Medical Subject Headings (MeSH) Arteriosclerosis [∗therapy]; Chelating Agents [∗therapeutic use]; Chelation Therapy [∗methods]; Edetic Acid [∗therapeutic use]; Ran- domized Controlled Trials as Topic

MeSH check words Humans

Chelation therapy for atherosclerotic cardiovascular disease (Review) 33 Copyright © 2009 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd. EDTA chelation therapy for cardiovascular disease: a systematic Review

Dugald MR Seely, Ping Wu and Edward J Mills

BMC Cardiovascular Disorders BioMed Central

Research article Open Access EDTA chelation therapy for cardiovascular disease: a systematic review Dugald MR Seely*1,2, Ping Wu2,3 and Edward J Mills4

Address: 1Canadian College of Naturopathic Medicine, Toronto, Canada, 2Institute of Medical Science, University of Toronto, Toronto, Canada, 3London School of Hygiene and Tropical medicine, University of London, UK and 4Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton, Canada Email: Dugald MR Seely* - [email protected]; Ping Wu - [email protected]; Edward J Mills - [email protected] * Corresponding author

Published: 01 November 2005 Received: 08 July 2005 Accepted: 01 November 2005 BMC Cardiovascular Disorders 2005, 5:32 doi:10.1186/1471-2261-5-32 This article is available from: http://www.biomedcentral.com/1471-2261/5/32 © 2005 Seely et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract Background: Numerous practitioners of both conventional and complementary and alternative medicine throughout North America and Europe claim that chelation therapy with EDTA is an effective means to both control and treat cardiovascular disease. These claims are controversial, and several randomized controlled trials have been completed dealing with this topic. To address this issue we conducted a systematic review to evaluate the best available evidence for the use of EDTA chelation therapy in the treatment of cardiovascular disease. Methods: We conducted a systematic review of 7 databases from inception to May 2005. Hand searches were conducted in review articles and in any of the trials found. Experts in the field were contacted and registries of clinical trials were searched for unpublished data. To be included in the final systematic review, the studies had to be randomized controlled clinical trials. Results: A total of seven articles were found assessing EDTA chelation for the treatment of cardiovascular disease. Two of these articles were subgroup analyses of one RCT that looked at different clinical outcomes. Of the remaining five studies, two smaller studies found a beneficial effect whereas the other three exhibited no benefit for cardiovascular disease from the use of EDTA chelation therapy. Adverse effects were rare but those of note included a few cases of hypocalcemia and a single case of increased creatinine in a patient on the EDTA intervention. Conclusion: The best available evidence does not support the therapeutic use of EDTA chelation therapy in the treatment of cardiovascular disease. Although not considered to be a highly invasive or harmful therapy, it is possible that the use of EDTA chelation therapy in lieu of proven therapy may result in causing indirect harm to the patient.

Background conventional therapy, there is no universally recognized Chelation therapy, a program of repeated intravenous standard protocol. Most protocols, however, share a administration of ethylene diamine tetraacetic acid degree of similarity. A typical protocol might consist of 30 (EDTA), often given in combination with vitamins and intravenously administered solutions of 3 grams of diso- minerals, has been touted as a safe alternative treatment dium EDTA with concomitant administration of varying for atherosclerotic vascular disease [1-3]. As this is a non- levels of ascorbic acid, B-vitamins, heparin, and the

Page 1 of 6 (page number not for citation purposes) BMC Cardiovascular Disorders 2005, 5:32 http://www.biomedcentral.com/1471-2261/5/32

minerals magnesium, copper, zinc, selenium and manga- pain free and maximal walking distances, subjective nese delivered over 1.5 to 3 hours in 500 ml to 1000 ml symptoms of angina, ankle brachial indices, digital sub- of normal saline. Therapy is often delivered on a weekly traction angiograms, transcutaneous oxygen tension, or biweekly basis and may be followed up with a less fre- blood cholesterol levels, and quality of life measures. We quent maintenance schedule. planned to conduct a meta-analysis, however, the limited number of trials, their clinical heterogeneity, and the var- There is a very strong market for this therapy with out-of- iability of outcomes made pooling impossible. We pocket costs for the use of EDTA to treat cardiovascular abstracted quality criteria data from each of the RCTs disease estimated to range from $400 million to $3 bil- assessed based on randomization, sample size determina- lion annually in the United States [4,5]. The use of chelation, dropouts, intention to treat, blinding, and allocation tion therapy for cardiovascular disease appears to concealment. contradict conventional medical thought, however, as three systematic reviews of clinical trials have concluded Results that chelation therapy is not supported by the evidence Main findings [1,2,6]. In contrast, one early meta-analysis of uncon- We found a total of fourteen studies for further analysis trolled trials and unpublished data claimed that EDTA [4,8-20]. Of these, we excluded seven, as they were either chelation therapy effectively improved the symptoms of review articles or non-randomized studies [8-14]. Seven cardiovascular disease in over 80% of cases [7]. studies of randomized controlled trials have been included in this systematic review [4,15-20]. Of the seven There is continued controversy as to the use of this therapy manuscripts, five were of distinct sample populations and a substantial amount of press given to it in popular [4,16-18,20], with three of the trials analyzing data from non-peer reviewed literature and on the Internet. Consid- only one participant sample [15,16,19]. Except for one ering the continued widespread usage and interest in trial published in Danish, all trials were published in Eng- EDTA chelation therapy we have endeavored to review the lish between 1963 and 2002. The data from the Danish most current state of the evidence in order to provide an trial was also published in English, however, thereby not update on this contentious and clinically relevant issue. requiring translation or inclusion. All included trials were conducted in the United States, Canada, Denmark, New Methods Zealand, or Germany. Search strategy We searched the following databases, from inception to Of the five sample populations tested, two studies (total n May 2005: MedLine, EMBASE, Cochrane controlled trials = 19) demonstrated a beneficial effect of EDTA chelation register (CENTRAL), AMED (Alternative Medicine);Alt therapy on cardiovascular disease measures and three HealthWatch;Pre-CINAHL;CINAHL;and the Nursing and (total n = 269) did not. Allied Health Collection. In addition we searched for unpublished and on-going trials through Clinicaltrials. Study characteristics and adverse effects gov and the National Research Register (UK). To be Since the most recent systematic review, a Cochrane included in this review, a trial had to be a randomized review, by Villaruz et al., 2002 [6], there has been one ran- controlled trial assessing EDTA in humans at risk for car- domized controlled clinical trial on the effect of EDTA on diovascular disease. We excluded non-randomized trials cardiovascular disease. The conclusion found by Villaruz and pharmacokinetic studies. No limits based on lan- et al was that 'there is insufficient evidence to decide on guage were imposed. Two reviewers (DS and PW) inde- the effectiveness or ineffectiveness of chelation therapy in pendently assessed the articles for inclusion and outcome improving clinical outcomes of patients with atheroscle- data. In the case of disagreement, arbitration was sought rotic cardiovascular disease' [6]. from EM. The most recent placebo-controlled trial conducted by A more detailed description of the search strategy used Knudtson and colleagues in 2002 [4] explored the effect and the results are presented [see additional file 1]. of chelation therapy on ischemic heart disease. When Authors of some of the trials were contacted to solicit their compared to placebo, a total of 33 rounds of EDTA treat- interpretation of the review and also to comment on any ment per patient was found to have no effect in any of the criticism that we had found in non-peer reviewed outcomes measured [see additional file 2]. Both groups literature. exhibited improvements from baseline but this was independent of EDTA use. Data abstraction We abstracted outcome data on ECG tests, exercise tests Two of the earliest randomized controlled trials, by Kitch- including treadmill tests, cycling time, time to ischemia, ell et al., 1963 [17] and Olszewer et al., 1990 [18], found

Page 2 of 6 (page number not for citation purposes) BMC Cardiovascular Disorders 2005, 5:32 http://www.biomedcentral.com/1471-2261/5/32

EDTA chelation therapy to have a beneficial effect on car- negative findings. As described above, the most recent diovascular risk profiles when compared to control. In the trial by Knudtson et al., 2002 did not find any evidence of first trial by Kitchell et al., 50% of the active group experi- efficacy in the treatment of 84 patients with coronary enced improved ECG readings after 6 and 12 weeks fol- artery disease through EDTA chelation therapy [4]. lowing the final treatment. This crossover trial was halted early because 3 of the 5 members in the placebo group We assessed for adverse effects amongst the trials and dropped out due to a lack of improvement. In the trial by found only a few cases of adverse events that might be Olszewer et al. (1990), the code was broken early because attributed to the EDTA [see additional file 2]. In the EDTA the exercise measures and the ankle brachial blood pres- treatment (vs. control) groups, one trial had a single (zero sure index all showed dramatic improvements in the che- controls) case of potential kidney toxicity in a population lation group [see additional file 2]. Of all the trials of 84 [4] and in another trial, eight cases (two controls) of analyzed, however, these studies used the least number of faintness and 12 cases of hypocalcemia (one control) in a participants, with sample sizes of 9 and 10 respectively. population of 153 were found [16]. No other adverse effects were noted in any of the RCTs. Since the two positive trials by Kitchell and Olszewer, three larger trials have been completed. None of the more Study quality recent trials has indicated any benefit from EDTA therapy In an assessment of the articles for quality criteria, two of on cardiovascular disease. One group of investigators, the trials did not describe randomization [15,17]; sample Guldager et al., 1992 [16], tested 153 patients with inter- size determination was only described in two trials [4,20]; mittent claudication for a number of clinical outcomes drop outs were excessive in one [17]; intention to treat [see additional file 2]. None of these outcomes demon- analysis was performed in four trials [4,18-20]; blinding strated evidence of improvement using parenteral EDTA was inadequate in two of the trials [17,18]; and allocation chelation. Two subsets of this initial sample were ana- concealment was only adequate in one trial [4]. A sum- lyzed separately; one for digital angiograms and transcu- mary of quality criteria findings is listed in table 1. The taneous oxygen profiles [19], and one for blood two trials with the highest quality [4,20] were the most cholesterol levels [15]. In neither of these outcomes was recent and, as discussed above, found no evidence of effi- there any evidence of improvement dependent the inter- cacy for EDTA chelation therapy in the treatment of vention alone. In the trial by Van Rij et al., 1994 [20], 32 cardiovascular disease. patients were tested for similar outcomes with the same

Table 1: Quality criteria of the randomized controlled trials assessed

Reference Randomization Sample size Drop outs Intention to Blinding Allocation determination treat analysis concealment

Kitchell 1963 [17] Not described Not described 33% of the No Allegedly Unclear (nor stated) treatment group participants upon cross-over blinded, but impossible to adequately assess Olszewer 1990 Described Not described None Yes Initially double Unclear [18] blinded but code was broken after 10 treatments and study was completed with single blind for remaining 10 treatments Sloth-Nielsen Described Not described None Yes Double Unclear 1991 [19] Guldager 1992 Described Not described 4 drop outs by 3 No Double: code Unclear [16] months; 30 drop broken at 3 outs by 6 months months Guldager 1993 Not described Not described Not mentioned No Double Unclear [15] Van Rij 1994 [20] Described Yes None Yes Double Unclear Knudtson 2002 [4] Described Yes 4 in placebo, 2 in Yes Double Well described treatment group

Page 3 of 6 (page number not for citation purposes) BMC Cardiovascular Disorders 2005, 5:32 http://www.biomedcentral.com/1471-2261/5/32

x 388 articles screened for inclusion

x 374 articles excluded as irrelevant or duplicate papers based on title and/or abstract

x 14 articles retrieved for further analysis

x 7 articles rejected due to being review, duplicate data, and/or ofnon- randomized trials

x 7 studies included in the systematic review (including data from 5 separate trials) x 2 studies analysed data from subpopulations of the larger study

Flow-chartFigure 1 detailing study selection and exclusion for systematic review Flow-chart detailing study selection and exclusion for systematic review.

Discussion Proposed mechanisms of action for the reversal of cardio- The overall evidence on EDTA chelation therapy argues vascular disease by EDTA include: calcium chelation against any clinical benefit with respect to cardiovascular resulting in dissolution of atheromatous plaques, free rad- disease. The evidence that we were able to find in support ical scavenging action, reduction of total body iron stores, of EDTA chelation for cardiovascular disease relies almost cell membrane stabilization, arterial dilatation due to cal- entirely on uncontrolled trials and a large body of anecdo- cium channel blocking action, improvement of arterial tal evidence. Given the parameters of the evidence so far, wall elasticity and increased production of nitric oxide and in keeping with patient values, it is encouraging that [4,21]. Critics have taken issue with some of these mech- the NIH is funding a large simple trial. A 5-year multi- anisms, however, claiming the use of outdated concepts center randomized placebo-controlled trial will hopefully on the pathophysiology of atherosclerosis and for deliver a conclusive answer on the question of whether or instance, the inability of EDTA, a water soluble com- not intravenous EDTA infusion has any role in the treat- pound, to effectively complex with plaque calcium [3]. ment of cardiovascular disease. Until such time, clinicians should openly discuss the use of complementary and Given the widespread usage of EDTA chelation therapy, alternative medicine and specifically chelation therapy an assessment of its safety is crucial. EDTA is responsible with patients in an informed and non-judgmental man- for a wide range of potential side effects including gas- ner. Clinicians should also discuss potential risks associ- trointestinal and musculoskeletal complaints, diaphore- ated with EDTA chelation therapy and the current lack of sis, fever, leukopenia, thrombocytopenia, kidney damage, evidence supporting it use in cardiovascular disease. mineral depletion, and hypocalcemia [19,22-24]. With

Page 4 of 6 (page number not for citation purposes) BMC Cardiovascular Disorders 2005, 5:32 http://www.biomedcentral.com/1471-2261/5/32

proper dose control and assessment of kidney function, by patients who use EDTA chelation therapy and the however, EDTA is not considered to be a particularly high- potential for harm associated with any intravenous inter- risk therapy and there is little doubt that it is safer than vention including the potential for adverse effects attrib- coronary bypass surgery. However, if EDTA has no efficacy utable directly to EDTA, clinicians should inquire about beyond placebo, then the possibility of adverse effects, the patient use and highlight the lack of evidence to support cost of treatment, and potential to avoid greater risk may its usage. well result in unjustifiable morbidity and mortality. List of abbreviations used There are several limitations to consider in our review. We CAM: complementary and alternative medicine cannot determine to what extent publication bias has affected our results. Empirical evidence has shown that ECG: electrocardiogram negative trials in CAM and other fields are less likely to be published [25]. Claims have been made however, that evi- EDTA: ethylene diamine tetraacetic acid dence in favor of EDTA chelation therapy may have been suppressed [26]. To mitigate as much as possible against NIH: National Institutes of Health these circumstances, we conducted thorough systematic searches of the literature. We contacted some, but not all RCT: randomized controlled trial of the authors of the studies to confirm our interpretation of the results. It may be that authors of original studies Competing interests excluded important information from the published The author(s) declare that they have no competing manuscripts in order to reduce the page length or to fol- interests. low reviewers' suggestions [27]. In many of the trials there was a significant improvement in outcome measures for Authors' contributions both control groups and chelation groups. This is a clear DS and EM conceived the design of the study, DS and PW indication of the need to conduct controlled trials in order conducted the searches and undertook the data extraction to pick up a type I error that might mistakenly attribute and analysis with assistance from EM. DS drafted the efficacy to the therapy in question. manuscript. EM edited the manuscript.

There is a large body of literature to support the use of Additional material EDTA in the treatment of cardiovascular disease; however, the vast majority of the literature relies on uncontrolled evidence. Supporters of EDTA chelation therapy have lev- Additional File 1 (text) EDTA and cardiovascular disease systematic review search strategy ied a number of criticisms against some of the RCTs employed. assessed in this review. These criticisms include but are Click here for file not limited to: too short of a treatment schedule, claims of [http://www.biomedcentral.com/content/supplementary/1471- incorrect statistical manipulation of the data, improper 2261-5-32-S1.doc] randomization, and high dropout rates [26,28-30]. The earliest trials incorporated 20 treatments of chelation Additional File 2 therapy and the American College for Advancement in (table) Randomized controlled trials assessing the therapeutic use of Medicine (ACAM), considered an authority in chelation EDTA for atherosclerotic cardiovascular disease. Click here for file therapy, claims that at least 30 treatments may be required [http://www.biomedcentral.com/content/supplementary/1471- for improvements to be noted. In the most recent study by 2261-5-32-S2.doc] Knudston et al., 2002, however, a protocol involving 33 treatment sessions were used and no positive therapeutic effect was found. In our review we included quality criteria and did not find evidence of any bias, however, we did Acknowledgements not obtain the raw data needed to conduct an independ- The Hospital for Sick Children Research Institute and the Sick Kids Foun- ent reanalysis. dation and for their support.

Conclusion References 1. Ernst E: Chelation therapy for coronary heart disease: An The findings of this systematic review should be of interest overview of all clinical investigations. Am Heart J 2000, to clinicians and patients alike. The use of EDTA by 140(1):139-141. patients as a treatment for cardiovascular disease and as 2. Ernst E: Chelation therapy for peripheral arterial occlusive disease: a systematic review. Circulation 1997, 96(3):1031-1033. an adjunct or alternative to surgery is not supported by the 3. Lewin MR: Chelation therapy for cardiovascular disease. highest quality of evidence. Considering the cost incurred Review and commentary. Tex Heart Inst J 1997, 24(2):81-89.

Page 5 of 6 (page number not for citation purposes) BMC Cardiovascular Disorders 2005, 5:32 http://www.biomedcentral.com/1471-2261/5/32

4. Knudtson ML, Wyse DG, Galbraith PD, Brant R, Hildebrand K, Pater- 28. Cranton EM: Negative Danish study of EDTA chelation biased. son D, Richardson D, Burkart C, Burgess E: Chelation therapy for Townsend Letter for Doctors 1992:604-605. ischemic heart disease: a randomized controlled trial. Jama 29. Olmstead SF: Critique of the Patch Chelation Therapy Trial. 2002, 287(4):481-486. 2002. 5. Lamas GA, Ackermann A: Clinical evaluation of chelation ther- 30. Chappell LT, Wilson J: Chelation therapy for vascular disease. apy: is there any wheat amidst the chaff? Am Heart J 2000, Circulation 1999, 99(1):164-165. 140(1):4-5. 6. Villarruz MV, Dans A, Tan F: Chelation therapy for atherosclerotic cardiovascular disease. Cochrane Database Syst Rev Pre-publication history 2002:CD002785. The pre-publication history for this paper can be accessed 7. Chappell LT, Stahl JP: The correlation between EDTA chelation here: therapy and improvement in cardiovascular function: a meta-analysis. J Adv Med 1993, 6:139-160. 8. Cheraskin E: Cardiovascular dynamics and EDTA chelation http://www.biomedcentral.com/1471-2261/5/32/prepub with multivitamin/trace mineral supplementation. J Orthomo- lecular Medicine 1991, 6(1):8-12. 9. Cranton EM, Misner B: Endurance performance gains following intravenous EDTA chelation therapy in a healthy 60-year-old athlete. Clin Pract Alternat Med 2001, 2(2):135-136. 10. Elihu N, Anandasbapathy S, Frishman WH: Chelation therapy in cardiovascular disease: ethylenediaminetetraacetic acid, deferoxamine, and dexrazoxane. J Clin Pharmacol 1998, 38(2):101-105. 11. Geurkink TL: Improved peripheral vascular function with low dose intravenous ethylene diamine tetraacetic acid (EDTA). Townsend Letter for Doctors & Patients 1994:722. 12. McDonagh EW, Rudolph CJ, Cheraskin E: An oculocerebrovascu- lometric analysis of the improvement in arterial stenosis following EDTA chelation therapy. J Holistic Med 1982, 4(1):21-23. 13. Rathmann KL, Golightly LK: Chelation therapy of atherosclerosis. Drug Intell Clin Pharm 1984, 18(12):1000-1003. 14. Schacter MB: EDTA Chelation Therapy - 2000. Total Health 2000, 22(4):42. 15. Guldager B, Brixen KT, Jorgensen SJ, Nielsen HK, Mosekilde L, Jelnes R: Effects of intravenous EDTA treatment on serum parathyroid hormone (1-84) and biochemical markers of bone turnover. Dan Med Bull 1993, 40(5):627-630. 16. Guldager B, Jelnes R, Jorgensen SJ, Nielsen JS, Klaerke A, Mogensen K, Larsen KE, Reimer E, Holm J, Ottesen S: EDTA treatment of intermittent claudication--a double-blind, placebo-controlled study. J Intern Med 1992, 231(3):261-267. 17. Kitchell JR, Palmon FJ, Aytan N, Meltzer LE: The treatment of coronary artery disease with disodium EDTA. A reappraisal. Am J Cardiol 1963, 11:501-506. 18. Olszewer E, Sabbag FC, Carter JP: A pilot double-blind study of sodium-magnesium EDTA in peripheral vascular disease. J Natl Med Assoc 1990, 82(3):173-177. 19. Sloth-Nielsen J, Guldager B, Mouritzen C, Lund EB, Egeblad M, Nor- regaard O, Jorgensen SJ, Jelnes R: Arteriographic findings in EDTA chelation therapy on peripheral arteriosclerosis. Am J Surg 1991, 162(2):122-125. 20. van Rij AM, Solomon C, Packer SG, Hopkins WG: Chelation therapy for intermittent claudication. A double-blind, randomized, controlled trial. Circulation 1994, 90(3):1194-1199. 21. Green DJ, O'Driscoll JG, Maiorana A, Scrimgeour NB, Weerasooriya R, Taylor RR: Effects of chelation with EDTA and vitamin B therapy on nitric oxide-related endothelial vasodilator function. Clin Exp Pharmacol Physiol 1999, 26(11):853-856. 22. Morgan BW, Kori S, Thomas JD: Adverse effects in 5 patients receiving EDTA at an outpatient chelation clinic. Vet Hum Toxicol 2002, 44(5):274-276. Publish with BioMed Central and every 23. Wirebaugh SR, Geraets DR: Apparent failure of edetic acid chelation therapy for the treatment of coronary atherosclerosis. scientist can read your work free of charge Dicp 1990, 24(1):22-25. "BioMed Central will be the most significant development for 24. Guldager B, Jorgensen PJ, Grandjean P: Metal excretion and mag- disseminating the results of biomedical research in our lifetime." nesium retention in patients with intermittent claudication treated with intravenous disodium EDTA. Clin Chem 1996, Sir Paul Nurse, Cancer Research UK 42(12):1938-1942. Your research papers will be: 25. Schmidt K, Pittler MH, Ernst E: Bias in alternative medicine is still rife but is diminishing. Bmj 2001, 323(7320):1071. available free of charge to the entire biomedical community 26. Cranton EM: Textbook on EDTA Chelation Therapy. 2nd edi- peer reviewed and published immediately upon acceptance tion. Charlottesville, Virginia , Hampton Roads Publishing Company Inc.; 2001. cited in PubMed and archived on PubMed Central 27. Roberts JC, Fletcher RH, Fletcher SW: Effects of peer review and yours — you keep the copyright editing on the readability of articles published in Annals of Internal Medicine. Jama 1994, 272(2):119-121. Submit your manuscript here: BioMedcentral http://www.biomedcentral.com/info/publishing_adv.asp

Page 6 of 6 (page number not for citation purposes) Should EDTA Chelation Therapy be Used Instead of Long-term Clopidogrel plus Aspirin to Treat Patients at Risk from Drug-Eluting Stents?

L.Terry Chappell

Alternative Medicine Review Volume 12, Number 2 2007 Review Article

Should EDTA Chelation Therapy be Used Instead of Long-term Clopidogrel plus Aspirin to Treat Patients at Risk from Drug-Eluting Stents?

L. Terry Chappell, MD

Abstract Risk of Thrombosis from Medicated The recently discovered increased risk of blood clots, leading to Stents myocardial infarction and sudden death beginning six months The use of drug-coated (drug-eluting) stents after medicated stents are implanted in patients following with percutaneous transluminal coronary angioplasty percutaneous transluminal coronary angioplasty (PTCA), has (PTCA) has come into widespread use in recent years. left cardiologists pondering what course of action to take. The Sirolimus-eluting stents were approved for use in April purpose of adding implanted medication to a stent is to prevent 2003, and 1.5 million have been implanted worldwide. thrombin accumulation and restenosis. However, these stents Paclitaxel-eluting stents were approved in March 2004, may increase, rather than decrease, the risk. Although long- and over three million have been implanted worldwide. term treatment with clopidogrel bisulfate (Plavix®) plus aspirin The purpose of these stents is to reduce restenosis. for at least 12 months has been suggested as a preventive Medicated stents have become increasingly popular; treatment, there is no evidence from randomized, controlled for example, they account for 90 percent of all stents sold in the United States in recent years. Although this trials that this treatment is effective for more than six months. percentage has decreased since reports of unfavorable Clopidogrel also increases the risk of major bleeding episodes. outcomes began appearing, medicated versions are still The author served as the primary investigator for a study that showed being widely used. cardiovascular patients treated with EDTA chelation therapy had An analysis of the Swedish Coronary Angi- a lower rate of subsequent cardiac events, including myocardial ography and Angioplasty Registry found a promising infarction and death, than those treated with cardiac medications, initial trend, with 13.4 fewer events per 1,000 patients PTCA, or coronary artery bypass graft (CABG). The data also treated during the first six months after the stents were indicated chelation therapy might be effective in preventing inserted.1 However, after six months, patients with thrombosis and cardiac events from stent implantation. drug-eluting stents had 12.7 more cardiac events per There is evidence EDTA chelation therapy might prevent thousand patients than those with bare metal stents, hypercoagulability resulting from the placement of stents, although not specifically medicated stents. Based on L. Terry Chappell, MD – Board-certified family physician; private practice in the limited data currently available, intravenous EDTA Bluffton, Ohio; President of the International College of Integrative Medicine.; advisor for the American Board of Clinical Metal Toxicology; member of the may be safe and effective for treating patients who have American College for Advancement in Medicine; clinical investigator for the NIH- funded Trial to Assess Chelation Therapy. implanted medicated stents. Prospective clinical trials Correspondence address: 122 Thurman Street, Bluffton, OH 45817. are needed, and EDTA should be included in those trials. Email: [email protected] (Altern Med Rev 2007;12(2):152-158)

Page 152 Copyright © 2007 Thorne Research, Inc. All Rights Reserved. No Reprint Without Written Permission. Alternative Medicine Review Volume 12, Number 2 June 2007 Alternative Medicine Review Volume 12, Number 2 2007 Review Article

due to an increase in thrombosis at the site of the stent. Journal of Medicine devoted virtually the entire March The risk of death increased by 0.5 percent per year and 8, 2007, issue to the subject. Unanswered questions re- the risk of myocardial infarction (MI) increased by 0.5- main because, depending on the techniques of analysis, 1.0 percent per year. At three years, the adjusted relative some studies concluded drug-eluting stents impose an risk for death was 1.32 for patients treated with medi- increased cardiovascular risk and higher mortality than cated stents compared to patients with non-medicated bare metal stents, while others detect no such increases stents. or even a reverse trend. However, all the studies con- Medicated stents have U.S. Food and Drug firmed the FDA conclusion that stents of any kind can Administration (FDA) approval for use in patients cause thrombosis and that the risk is particularly high with single-vessel blockage (no more than two stents with drug-eluting stents used for off-label indications. per patient) and no other significant medical problems. In September 2006, an FDA panel concluded that pa- Mechanism and Evidence for the Use of tients who have drug-eluting stents have a significantly Clopidogrel for Coronary Artery Disease increased risk for developing blood clots. The report also Clopidogrel inhibits adenosine-induced plate- noted studies indicating a small but significant increased let aggregation. The evidence for its use for cardiovas- risk for myocardial infarction and stroke.2 Drug-eluting cular disease is primarily based on two randomized, stent patients had a lesser need for reintervention; how- controlled clinical trials. Aspirin inhibits thrombox- ever, long-term survival and myocardial infarction-free ane A2-induced platelet aggregation. Thus, combining survival were not improved, compared to patients with clopidogrel with aspirin results in two slightly differ- bare metal stents. ent mechanisms of action. The two drugs together are These risks are of great concern, but an even thought to be more effective than either drug alone. greater concern, according to the FDA panel, is that at The Clopidogrel versus Aspirin in Patients at least 60 percent of coated-stent use was off-label, due Risk of Ischemic Events (CAPRIE) trial,5 compared as- to the simultaneous insertion of more than two stents pirin with clopidogrel for an average time of 1.6 years in and/or the implantation of coated stents in patients 19,185 patients with established vascular disease. The with concomitant medical problems, such as myocardial incidence of new ischemic strokes was 4.6 percent with infarction, multivessel disease, and diabetes.3 For indi- clopidogrel and 4.8 percent with aspirin. The incidence viduals treated for an off-label indication, the drug-elut- of new myocardial infarctions was 2.9 percent with ing stent patients had higher rates of death from cardio- clopidogrel and 3.5 percent with aspirin. The absolute vascular causes, more nonfatal myocardial infarctions, risk was lowered only 0.2 and 0.6 percent, respectively, and more stent thromboses than bare metal stent pa- and the vascular death rates were identical. The relative tients. One reason for the expanded use of coated stents risks showed a preference for clopidogrel, but based on is that Medicare pays for these off-label, non-FDA-ap- death rates, the trial did not actually show clopidogrel proved indications. was superior to aspirin for this broad group of patients. The Occluded Artery Trial (OAT) added The Clopidogrel in Unstable angina to prevent more confusion by challenging the basis of the open Recurrent Events (CURE) trial of 12,562 patients with artery theory.4 It looked at 2,166 stable patients with acute coronary syndrome compared clopidogrel plus totally occluded arteries several days after an MI and aspirin to aspirin alone over a one-year period.6 Add- compared those treated with angioplasty plus stent and ing clopidogrel lowered the cardiovascular death rate medications to those treated with medications alone. from 5.5 to 5.1 percent, ischemic stroke incidence from The patients treated with medications alone did slightly 1.4 to 1.2 percent, and MI incidence from 6.6 to 5.2 better than those treated with stents, whether the stent percent. A small benefit was attributed to combining was drug-eluting or not. The difference was statistically the two drugs, but this benefit was demonstrated only significant. for patients with the specific vascular problem of acute The dilemma about when to use medicated coronary syndrome (either unstable angina or non-ST stents and how to treat patients who have had them elevation myocardial infarction). Importantly, for pa- inserted has been so controversial the New England tients who had acute coronary syndrome, the benefit

Page 153 Copyright © 2007 Thorne Research, Inc. All Rights Reserved. No Reprint Without Written Permission. Alternative Medicine Review Volume 12, Number 2 June 2007 Alternative Medicine Review Volume 12, Number 2 2007 EDTA Chelation Therapy

date does not support this therapy, and there is a Table 1. EDTA Cardiac Events Study: 220 Patients with significant risk of brain hemorrhage. Three-year Follow-up Mechanism and Evidence for the Predicted Actual Use of Intravenous EDTA for Coronary Artery Disease Myocardial infarctions 15 0 Many studies affirm the usefulness of Deaths 6 0 intravenous (I.V.) EDTA8,9 in the treatment of vascular disease, while others report negative Subsequent angioplastiesangioplasties 31 2 conclusions.10 None of the studies meet the cri- Subsequent CABG 16 6 teria of two National Institutes of Health (NIH) review panels for definitive proof. All have been under-powered and many studies have been uncontrolled, poorly randomized, and selectively occurred in the first two months after the syndrome was reported. The NIH has funded the Trial to As- diagnosed. The CURE trial limited stroke analysis to is- sess Chelation Therapy (TACT), which is in progress 1chemic strokes; apparently patients who suffered hem- at more than 100 centers in the United States and Can- orrhagic strokes were outliers removed from the study. ada. This randomized, controlled trial is designed to de- The CURE study demonstrated clopidogrel termine whether patients treated with EDTA chelation had no impact in reducing the number of patients who therapy in addition to standard medications and who subsequently needed a coronary artery bypass graft have had a previous MI will have fewer cardiac events (CABG) or PTCA (with or without stenting). There and deaths than others treated with standard medica- was no reduction in cardiac deaths for patients with tions alone. demonstrated vascular disease in general, and for those Anticoagulation is a common laboratory use with acute coronary syndrome the benefit was marginal for and known mechanism of EDTA. Studies demon- and limited to the initial two months of therapy. strate EDTA inhibits platelet aggregation induced by Eisenstein et al conducted an observational adenosine, epinephrine, and thrombin, while preserving study of patients who received angioplasty and stents at collagen-induced aggregation.11,12 Thus, EDTA pro- 7 the Duke Heart Center from 2000-2005. They found vides good therapeutic effect, with three mechanisms drug-eluting stent patients who discontinued clopido- that reduce platelet aggregation, while it maintains a grel after six months experienced an increased rate of safety factor by not inhibiting collagen-induced aggre- death and non-fatal MI compared to those treated for gation. Clopidogrel inhibits only adenosine-induced 12 months. For patients with bare metal stents there aggregation. was no such increase. The authors concluded there was EDTA is approved for chelating lead from the a beneficial trend for using clopidogrel for 12 months, body. Recent studies have shown even small amounts but that a randomized clinical trial of sufficient power of lead can be a major risk for all-cause mortality, car- was needed to provide evidence to determine the opti- diovascular mortality, and myocardial infarction.13 Lead mal length of treatment with clopidogrel. The research- has also been shown to increase free radical activity.14 ers did not analyze bleeding complications. Thus, documented mechanisms of action support the The increased risk of blood clots from drug- use of EDTA for coronary artery disease, whether a eluting stents does not begin until six months after in- stent is also used or not. sertion. Despite the lack of major clinical benefit and The author of this article was the chief inves- the existence of a significant risk for major bleeding tigator for a multicenter, retrospective study of 220 pa- episodes, many cardiologists treat patients who have tients with known vascular disease treated from 1992- medicated stents with clopidogrel plus aspirin for at 2001 with I.V. EDTA and followed for three years.15 least 12 months, if not indefinitely. Clinical evidence to The purpose of the study was to compare whether

Page 154 Copyright © 2007 Thorne Research, Inc. All Rights Reserved. No Reprint Without Written Permission. Alternative Medicine Review Volume 12, Number 2 June 2007 Alternative Medicine Review Volume 12, Number 2 2007 Review Article

cardiac events and death rates were reduced compared chelation group was 72-percent male; the PTCA groups to what has been reported in the literature for stan- were 82-percent male. Forty-eight percent of the chela- dard therapy (Table 1). Study patients were compared tion group had never smoked compared to 42 percent of to similar groups of patients reported in the literature the PTCA groups. treated primarily with PTCA, CABG, or conventional Of the groups treated primarily with PTCA, cardiac medications; both surgical groups were also 7.3 percent had an MI within three years of follow-up, treated with cardiac medications as indicated. while the incidence of death from any cause was 3.2 percent in the PTCA groups. Of those treated with PTCA, 22.3 percent Table 2. Surgical Procedures Required for Patients Treated Primarily needed another angioplasty within with Angioplasty and Stents with Three-year Follow-up three years, and 11.8 percent underwent a CABG procedure. Thus, Control Groups EDTA Treated 34.1 percent of patients treated with (25 patients) angioplasty required another surgical procedure within three years. Percent repeat angioplasties 22.3% 4% These statistics are comparable to other reports from the literature. Percent CABG required 11.8% 0% Twenty-five of 220 patients in the chelation study group had been treated previously with a PTCA, mostly with stent place- Rakesh Shukla of the Center for Biostatistical ment (Table 2). Although this is not a large group, the Services at the University of Cincinnati Medical School 25 patients did remarkably well, as only one required performed a meta-analysis on the groups treated with a repeat PTCA within the three-year follow-up. None conventional therapies and determined the rate of sub- of them had an MI or a CABG procedure, and none of sequent cardiac events (MI, CABG, and PTCA) and them died. The data from the literature indicate these death together and separately. From the rate of cardiac outcome numbers are less than expected from a group events and deaths in the meta-analysis, he predicted of patients treated primarily with PTCA and followed the 220 patients treated with EDTA chelation thera- with conventional cardiac medications. py should have suffered 15 MIs and six deaths during The chelation study group reported no ad- the three-year follow-up period. There were no MIs verse clinical effects from the intravenous use of EDTA, and no deaths in the patients treated with EDTA che- which was administered according to the published lation therapy. Furthermore, it was predicted by the protocol,15 confirming many reports from the literature meta-analysis that there should have been 16 CABG that EDTA is a safe treatment when used properly.9,15 procedures and 31 PTCAs for the comparable group Major bleeding episodes have not been a concern with a of 220 patients in the chelation study. There were only properly administered EDTA protocol. six CABG procedures and two PTCAs in the EDTA According to the published protocol, EDTA treatment group. is administered with specified additives intravenously This article focuses on the groups of patients once or twice weekly for 20-30 treatments, each lasting treated primarily with angioplasty with or without 1.5-3 hours, depending on the dose used. The recom- stents. A literature search found seven groups of patients mended dose of disodium magnesium EDTA is 50 mg/ treated primarily with PTCA in a time period similar kg. Some clinicians give a maximum dose of 1.5 g, oth- to the study group. The average number of patients for ers 3.0 g; the dose is adjusted based on kidney function. the PTCA groups was 226. Other characteristics of the For maintenance, treatments are administered monthly PTCA groups were similar to the EDTA group. The because new platelets are formed approximately every average age of the EDTA group was 64 years, while 3.5 weeks; monthly treatments provide continued plate- the average age of the PTCA group was 58 years. The let aggregation inhibition.

Page 155 Copyright © 2007 Thorne Research, Inc. All Rights Reserved. No Reprint Without Written Permission. Alternative Medicine Review Volume 12, Number 2 June 2007 Alternative Medicine Review Volume 12, Number 2 2007 EDTA Chelation Therapy

The Case for Using EDTA Chelation found in patients, pose a substantial risk for cardiovas- Therapy to Prevent Activated Clotting cular disease. EDTA removes lead from the body. What is needed is clinical data to confirm the assertion that in Patients Treated Previously with chelation therapy might be a useful treatment. Medicated Stents Although data from a prospective, controlled The mechanism of action of I.V. EDTA argu- trial does not exist, the data from the EDTA cardiac ably fits well as an agent to prevent clotting induced by events study is discussed above. Twenty-five out of 220 a medicated stent. It could be theorized its inhibition of patients treated with EDTA were treated previously adenosine-, epinephrine-, and thrombin-induced plate- with PTCA with or without stent placement. None of let aggregation might be more effective than mecha- these patients had a myocardial infarction or ischemic nisms that inhibit platelet aggregation more narrowly, stroke, and none of them died during the three-year as is the case of clopidogrel, which inhibits only adenos- follow-up period. Only one of the 25 patients required ine-induced aggregation. Low levels of lead, commonly a repeat PTCA, and no CABG procedures were performed during the follow-up period, indicating the incidence Table 3. Thrombosis from Drug-eluting Stents Compared to EDTA of cardiac events was apparently significantly reduced by treat- EDTA Long-term Clopidogrel ment with EDTA from what would be expected in a group FDA indication No No of patients with coronary artery disease treated with PTCA and Clinical trials No No stents. Platelet inhibition 3 Mechanisms 1 Mechanism Table 3 compares the albeit limited data on intrave- Risk of major bleeding Minimal Yes nous EDTA to data supporting the long-term use of clopidogrel Preliminary data showing Yes Yes plus aspirin to prevent clot- clinical effectiveness ting in the presence of medicated stents and the effect of

Table 4. Optimal Treatment for Coronary Artery Disease

Conventional Therapy Alternative Therapy Optimize lifestyle Yes Yes Platelet aggregation Clopidogrel; Aspirin EDTA; Fish oils Lipids “Statin” drug Red yeast rice; Niacin; Fish oils Inflammation “Statin” drug Red yeast rice; Fish oils Arrhythmias Beta-blocker Magnesium Blood pressure Beta-blocker; ACE Magnesium; Fish oils; Garlic; inhibitor, Diuretic Medication if necessary Removal of lead None EDTA chelation therapy

Page 156 Copyright © 2007 Thorne Research, Inc. All Rights Reserved. No Reprint Without Written Permission. Alternative Medicine Review Volume 12, Number 2 June 2007 Alternative Medicine Review Volume 12, Number 2 2007 Review Article

clopidogrel in preventing cardiac events for other con- Intravenous EDTA has a documented mecha- ditions, the latter of which demonstrated virtually no nism of action. There is reliable evidence from the lit- reduction in death rates. Both therapies are being used erature that it may prevent clotting and subsequent car- off-label in the treatment of coronary artery disease. diac events, including premature death, in patients who Safety is an important factor in choosing which have had PTCA with or without stent placement. The medication to use. Intravenous disodium magnesium evidence is not conclusive, but it compares favorably to EDTA has been used safely since the modern protocol the body of evidence supporting the use of clopidogrel. was established 35 years ago. Clopidogrel carries a sig- Neither treatment has an FDA indication for long- nificant risk of hemorrhage. term use to prevent complications from the placement Clopidogrel is often given in a loading dose of medicated stents. The data in hand, however, appears followed by 75 mg daily. EDTA is usually given intra- to indicate that chelation therapy may be an effective venously once or twice per week for 20-30 treatments, component in a comprehensive program to lesson the followed by monthly treatments. The cost of the initial clotting risk from drug-eluting stents. In addition, any course of treatment is higher with EDTA, but the cost cardiovascular rehabilitation program should include of maintenance treatment with EDTA is probably less aggressive lifestyle change. While there is some evidence than ongoing treatment with clopidogrel. Furthermore, that treatment with clopidogrel might have benefit dur- patients receiving monthly I.V. treatments are moni- ing the 3-6 months following the placement of a medi- tored more closely than those given a three-month pre- cated stent, long-term use of clopidogrel has no proven scription for clopidogrel with refills up to one year. At benefit. EDTA appears to be safer than clopidogrel be- such visits, lifestyle changes can be reinforced, resulting cause it is less likely to cause major bleeding episodes. in better compliance, an important consideration given In a March 8, 2007, New England Journal of adherence to multiple healthy lifestyle factors might be Medicine article, the FDA called for randomized, con- more important than any therapeutic drug in prevent- trolled trials to determine the best treatment strategies ing cardiovascular deaths.16 An optimal integrative ap- for patients with complex cardiovascular conditions, proach to prevention and treatment of coronary artery such as diabetes, acute myocardial infarction, and mul- disease is summarized in Table 4. tivessel coronary artery disease.17 The FDA emphasized that safety end points such as death and myocardial in- Conclusion farction should be studied, and the duration of therapy No scientific evidence supports long-term use with clopidogrel and aspirin needs to be determined. of clopidogrel in preventing potentially fatal clotting This author proposes treatment with EDTA chela- linked to medicated stents. Quite the contrary, there tion therapy should also be studied for use in patients is some indication that the long-term use of clopido- treated with medicated stents because of its safety and grel might put the patient at risk for a major bleeding possible effectiveness. In the article, the FDA did not episode, especially a hemorrhagic stroke. The benefit of address the concern that patients with occluded arteries using clopidogrel appears to be limited to the first 60 treated with medications alone might have better out- days after diagnosing acute coronary syndrome, which comes than those treated with angioplasty and stents. applies to a minority of patients who have had stents The limited data available indicates EDTA inserted. There is also some evidence that clopidogrel may be safer and more effective than clopidogrel in pre- plus aspirin inhibits stent thrombosis during the first venting clots, myocardial infarctions, additional surgical 3-6 months after stent placement. The recently detected procedures, and death as complications from medicated increased risk of clotting from medicated stents occurs stents placed in coronary arteries, particularly in pa- later, at least six months after insertion. Thus, there is tients with complex conditions. no proof via randomized, clinical trials that treating EDTA has the additional benefit of removing patients with a drug-eluting stent in conjunction with low levels of lead that add a major risk of cardiovascu- clopidogrel for more than six months is helpful or safe. lar disease and increased mortality. EDTA chelation therapy is being examined in the TACT study. Prelimi- nary data from the study cited above, which looked at

Page 157 Copyright © 2007 Thorne Research, Inc. All Rights Reserved. No Reprint Without Written Permission. Alternative Medicine Review Volume 12, Number 2 June 2007 Alternative Medicine Review Volume 12, Number 2 2007 EDTA Chelation Therapy

the same end points as TACT, indicate EDTA might 8. Chappell LT, Stahl JP. The correlation between be effective in preventing cardiac events for patients EDTA chelation therapy and improvement in with known vascular disease. Since the use of angio- cardiovascular function: a meta-analysis. J Adv Med 1993;6:139-163. plasty and stents does not appear to improve outcomes 9. Olmstead SF. A Critical Review of EDTA Chelation in patients with totally occluded arteries, the addition Therapy in the Treatment of Occlusive Vascular Disease. of EDTA to standard medications may be a promising Klamath Falls, OR: Merle West Medical Center treatment for coronary artery disease, whether or not Foundaton; 1998. drug-eluting stents have been implanted. 10. Knudtson ML, Wyse DG, Galbraith PD, et al. Chelation therapy for ischemic heart disease: a randomized controlled trial. JAMA 2002;287:481- Disclosure 486. The author has no financial interest in EDTA 11. Kindness G, Frackelton JP. Effect of ethylene diamine other than its use for patients in his own office. tetraacetic acid (EDTA) on platelet aggregation in human blood. J Adv Med 1989:2:519-530. 12. Zucker MB, Grant RA. Nonreversible loss of platelet References aggregability induced by calcium deprivation. Blood 1. Lagerqvist B, James SK, Stenestrand U, et.al. 1978;52:505-513. Long-term outcomes with drug-eluting stents 13. Menke A, Muntner P, Batuman V, et al. Blood versus bare-metal stents in Sweden. N Engl J Med lead below 0.48 micromol/L (10 microg/dL) 2007;356:1009-1019. and mortality among US adults. Circulation 2. http://www.fda.gov/cdrh/news/091406.html 2006;114:1388-1394. [Accessed May 14, 2007] 14. Cranton EM. A Textbook on EDTA Chelation 3. http://www.fda.gov/cdrh/news/010407.html Therapy. 2nd ed. Charlottesville, VA: Hampton [Accessed May 14, 2007] Roads Publishing Company; 2001. 4. Hochman JS, Lamas GA, Buller CE, et al. Coronary 15. Chappell LT, Shukla R, Yang J, et al. Subsequent intervention for persistent occlusion after myocardial cardiac and stroke events in patients with known infarction. N Engl J Med 2006;355:2395-2407. vascular disease treated with EDTA chelation 5. No authors listed. A randomised, blinded, trial therapy. Evid Based Integr Med 2005;2:27-35. of clopidogrel versus aspirin in patients at risk of 16. Chiuve SE, McCullough ML, Sacks FM, et al. ischaemic events (CAPRIE). CAPRIE Steering Healthy lifestyle factors in the primary prevention Committee. Lancet 1996;348:1329-1339. of coronary heart disease among men: benefits 6. Yusuf S, Zhao F, Mehta SR, et al. Effects of among users and nonusers of lipid-lowering clopidogrel in addition to aspirin in patients with and antihypertensive medications. Circulation acute coronary syndromes without ST-segment 2006;114:160-167. elevation. N Engl J Med 2001;345:494-502. 17. Farb A, Boam AB. Stent thrombosis redux – the 7. Eisenstein EL, Anstrom KJ, Kong DF, et al. FDA perspective. N Engl J Med 2007;356:984-987. Clopidogrel use and long-term clinical outcomes after drug-eluting stent implantation. JAMA 2007;297:159-168.