Vitamin K for improved anticoagulation control in patients receiving warfarin

  • Review
  • Intervention

Authors


Abstract

Background

Effective use of warfarin involves keeping the international normalised ratio (INR) within a relatively narrow therapeutic range. However, patients respond widely to their dose of warfarin. Overcoagulation can lead to an increased risk of excessive bleeding, while undercoagulation can lead to increased clot formation. There is some evidence that patients with a variable response to warfarin may benefit from a concomitant low dose of vitamin K.

Objectives

To assess the effects of concomitant supplementation of low-dose oral vitamin K for anticoagulation control in patients being initiated on or taking a maintenance dose of warfarin.

Search methods

To identify previous reviews, we searched the Database of Abstracts of Reviews of Effects (DARE via The Cochrane Library, Wiley) (Issue 2, 2011). To identify primary studies, we searched the Cochrane Central Register of Controlled Trials (CENTRAL via The Cochrane Library, Wiley) (Issue 2, 2014), Ovid MEDLINE (R) In-Process & Other Non-Indexed Citations database and Ovid MEDLINE (R) (OvidSP) (1946 to 25 February 2014), Embase (OvidSP) (1974 to week 8 of 2014), Science Citation Index Expanded™ & Conference Proceedings Citation Index - Science (Web of Science™) (1945 to 27 February 2014), and the NHS Economics Evaluations Database (NHS EED) (via The Cochrane Library, Wiley) (Issue 2, 2014). We did not apply any language or date restrictions. We used additional methods to identify grey literature and ongoing studies.

Selection criteria

Randomised controlled trials comparing the addition of vitamin K versus placebo in patients initiating warfarin or already taking warfarin.

Data collection and analysis

Two review authors independently selected and extracted data from included studies. When disagreement arose, a third author helped reached a consensus. We also assessed risk of bias.

Main results

We identified two studies with a total of 100 participants for inclusion in the review. We found the overall risk of bias to be unclear in a number of domains. Neither study reported the time taken to the first INR in range. Only one study (70 participants) reported the mean time in therapeutic range as a percentage. This study found that in the group of participants deemed to have poor INR control, the addition of 150 micrograms (mcg) oral vitamin K significantly improved anticoagulation control in those with unexplained instability of response to warfarin. The second study (30 participants) reported the effect of 175 mcg oral vitamin K versus placebo on participants with high variability in their INR levels. The study concluded that vitamin K supplementation did not significantly improve the stability of anticoagulation for participants on chronic anticoagulation therapy. However, the study was only available in abstract form, and communication with the lead author confirmed that there were no further publications. Therefore, we interpreted this conclusion with caution. Neither study reported any thromboembolic events, haemorrhage, or death from the addition of vitamin K supplementation.

Authors' conclusions

Two included studies in this review compared whether the addition of a low dose (150 to 175 mcg) of vitamin K given to participants with a high-variability response to warfarin improved their INR control. One study demonstrated a significant improvement, while another smaller study (published in abstract only) suggested no overall benefit. Currently, there are insufficient data to suggest an overall benefit. Larger, higher quality trials are needed to examine if low-dose vitamin K improves INR control in those starting or already taking warfarin.

Résumé scientifique

La vitamine K pour un meilleur contrôle de l'anticoagulation chez les patients recevant de la warfarine

Contexte

Une utilisation efficace de la warfarine consiste à maintenir le rapport international normalisé (RIN) dans une marge thérapeutique relativement étroite. Cependant, les patients répondent largement à leur dose de warfarine. Une forte coagulation peut conduire à un risque accru de saignements excessifs, tandis qu'une faible coagulation peut entraîner une augmentation de la formation de caillots. Certaines preuves indiquent que les patients avec une réponse variable à la warfarine pourraient bénéficier de l'administration concomitante d'une faible dose de vitamine K.

Objectifs

Évaluer les effets de la supplémentation concomitante par voie orale à faible dose de vitamine K pour le contrôle de l'anticoagulation chez les patients commençant ou prenant une dose d'entretien de warfarine.

Stratégie de recherche documentaire

Afin d'identifier les revues précédentes, nous avons effectué des recherches dans Database of Abstracts of Reviews of Effects (DARE via la Bibliothèque Cochrane, Wiley) (numéro 2, 2011). Pour identifier des études primaires, nous avons effectué des recherches dans le registre Cochrane des essais contrôlés (CENTRAL via la Bibliothèque Cochrane, Wiley) (numéro 2, 2014), Ovid MEDLINE (R) en cours et autres citations non indexées et Ovid MEDLINE (R) (OvidSP) (de 1946 au 25 février 2014), Embase (OvidSP) (de 1974 à la semaine 8 de 2014), Science Citation Index Expanded" & Conference Proceedings Citation Index - Science (Web of Science") (de 1945 au 27 février 2014) et la base de données d'évaluation économique du NHS (NHS EED) (via la Bibliothèque Cochrane, Wiley) (numéro 2, 2014). Aucune restriction concernant la langue ou la date n'a été appliquée. Nous avons utilisé d'autres méthodes pour identifier la littérature grise et les études en cours.

Critères de sélection

Essais contrôlés randomisés comparant l'ajout de vitamine K par rapport à un placebo chez les patients qui commencent ou prennent déjà de la warfarine.

Recueil et analyse des données

Deux auteurs de la revue ont indépendamment sélectionné et extrait les données des études incluses. En cas de désaccords, un troisième auteur a aidé à trouver un consensus. Nous avons également évalué les risques de biais.

Résultats principaux

Nous avons identifié deux études portant sur un total de 100 participants pour l'inclusion dans la revue. Le risque global de biais était incertain dans plusieurs domaines. Aucune étude n'a rapporté le temps nécessaire au première RIN en marge. Une seule étude (70 participants) a rapporté le temps moyen de la marge thérapeutique en pourcentage. Cette étude a constaté que dans le groupe de participants jugés avoir un contrôle inadéquat du RIN, l'ajout de 150 microgrammes (mcg) de vitamine K par voie orale améliorait significativement le contrôle de l'anticoagulation chez les patients répondant à la warfarine par une instabilité inexpliquée. La deuxième étude (30 participants) a rapporté l'effet de 175 mcg de vitamine K par voie orale par rapport à un placebo chez des participants avec une variabilité élevée dans leurs niveaux du RIN. L'étude a conclu que la supplémentation en vitamine K n'améliorait pas significativement la stabilité de l'anticoagulation chez les participants sous traitement chronique par anticoagulants. Cependant, l'étude n'était disponible que sous la forme de résumé et une communication avec l'auteur principal a confirmé qu'il n'y avait pas d'autres publications. Par conséquent, nous avons interprété cette conclusion avec prudence. Aucune étude ne rapportait d'événements thromboemboliques, d'hémorragie ou de décès suite à l'ajout de supplémentation en vitamine K.

Conclusions des auteurs

Deux études incluses dans cette revue comparaient si l'ajout d'une dose faible (150 à 175 mcg) de vitamine K administrée aux participants avec une grande variabilité de réponse à la warfarine améliorait le contrôle de leur RIN. Une étude a démontré une amélioration significative, tandis qu'une autre étude plus petite (publiée uniquement sous forme de résumé) ne suggérait aucun bénéfice global. Actuellement, il n'existe pas suffisamment de données permettant de suggérer un bénéfice global. Des essais de plus grande taille et de meilleure qualité sont nécessaires pour examiner si la vitamine K à faible dose améliore le contrôle du RIN chez les patients commençant ou prenant déjà de la warfarine.

アブストラクト

ワルファリン投与患者における抗凝固管理改善のためのビタミンK

背景

ワルファリンの効果的使用には、国際標準比(INR)を比較的狭い治療域内に保つことが伴う。しかし、患者は、自身の用量のワルファリンに対し幅広い反応を示す。過剰凝固管理は過度の出血リスク上昇につながり得るが、その一方、凝固管理不良は血栓形成のリスクにつながり得る。ワルファリンに対する反応が不安定な患者は低用量ビタミンK併用から利益を得られる可能性があるという一定程度のエビデンスがある。

目的

ワルファリンを開始するまたは維持用量を投与中の患者における抗凝固コントロールに対する低用量ビタミンK補充併用の効果を評価すること。

検索戦略

過去のレビューを同定するために、Database of Abstracts of Reviews of Effects(DARE via The Cochrane Library, Wiley)(2011年第2号)を検索した。 主要な試験を同定するために、Cochrane Central Register of Controlled Trials(CENTRAL via he Cochrane Library, Wiley)( 2014年第2 号)Ovid MEDLINE (R) In-Process & Other Non-Indexed Citations database and Ovid MEDLINE(R)(OvidSP)(1946 年~ 2014年2月25日)、Science Citation Index Expanded™ & Conference Proceedings Citation Index – Science(Web of Science™)(1945年~2014年2月27日)、NHS Economics Evaluations Database(NHS EED)(via The Cochrane Library, Wiley)(2014年第2号)を検索した。 言語、日付による制限を適用しなかった。 灰色文献および進行中の試験を同定するためにその後追加された方法を用いた。

選択基準

ワルファリンを開始するまたは既に投与中の患者を対象としてビタミンK追加をプラセボと比べたランダム化比較試験。

データ収集と分析

2名のレビュー著者が独立して選択し、選択した試験からデータを抽出した。不一致があれば第三の著者が援助して合意に至った。バイアスのリスクも評価した。

主な結果

本レビューのために試験2件(参加者100例)を同定した。全体としてのバイアスのリスクは多数の領域で不明であるとわかった。いずれの試験も初回INR範囲内までの時間を報告していなかった。 1件のみ(参加者70例)が治療域内平均時間をパーセンテージとして報告していた。この試験では、INRコントロールが不良と考えられる群の患者においては、経口ビタミンK 150 mcg/日追加によりワルファリンに対し原因不明の不安定を示す患者で抗凝固コントロールが有意に改善したことがわかった。この第二の試験(参加者30例)では、INRレベルが非常に不安定な参加者を対象として経口ビタミンK 175 mcgをプラセボと比べた。この試験では、長期抗凝固療法中の参加者においてビタミンK補充により抗凝固の安定性は有意に改善しなかったと結論された。しかし、この試験は抄録形式でしか利用できず、筆頭著者に問い合わせ、他に公表物はないことを確認した。そのため、この結論を慎重に解釈した。いずれの試験でもビタミンK補充による血栓塞栓イベント、出血、死亡は報告されていなかった。

著者の結論

選択された試験2件では、ワルファリンに対する反応が非常に不安定な参加者に追加で低用量ビタミンK(150 ~175 mcg)を投与したところINRコントロールが改善した。1件では、有意な改善が示されたが、もう1件のより小規模な試験(抄録のみの発表)では全体的利益は示唆されなかった。現在、全体的利益を示唆するためにはデータが不十分である。ワルファリンを開始するまたは既に投与中の患者を対象として低用量ビタミンKによりINRコントロールが改善するかどうかを検討するためにはより大規模で質の高い試験が必要である。

Resumen

Vitamina K para mejorar el control de la anticoagulación en pacientes que reciben warfarina

Antecedentes

El uso efectivo de warfarina incluye el mantenimiento de la razón internacional normalizada (RIN) dentro de un rango terapéutico relativamente estrecho. Sin embargo, los pacientes responden de forma variable a la dosis de warfarina. El exceso de coagulación puede dar lugar a un mayor riesgo de hemorragia excesiva, mientras que la falta de coagulación puede resultar en una mayor formación de coágulos. Hay algunas pruebas de que los pacientes con una respuesta variable a la warfarina pueden beneficiarse con una dosis baja concomitante de vitamina K.

Objetivos

Evaluar los efectos de la administración concomitante de suplementos de vitamina K oral en dosis baja para el control de la anticoagulación en pacientes que inician o que reciben una dosis de mantenimiento de warfarina.

Métodos de búsqueda

Para identificar revisiones anteriores, se hicieron búsquedas en la Database of Abstracts of Reviews of Effects (DARE vía The Cochrane Library, Wiley) (número 2, 2011). Para identificar estudios primarios, se hicieron búsquedas en el Registro Cochrane Central de Ensayos Controlados (Cochrane Central Register of Controlled Trials) (CENTRAL vía The Cochrane Library, Wiley) (número 2, 2014), Ovid MEDLINE (R) In-Process & Other Non-Indexed Citations database y Ovid MEDLINE (R) (OvidSP) (1946 hasta el 25 febrero 2014), Embase (OvidSP) (1974 hasta la semana 8 de 2014), Science Citation Index Expanded™ & Conference Proceedings Citation Index - Science (Web of Science™) (1945 hasta el 27 febrero 2014) y en la NHS Economics Evaluations Database (NHS EED) (vía The Cochrane Library, Wiley) (número 2, 2014). No se aplicó ninguna restricción de idioma o de fecha. Se utilizaron métodos adicionales para identificar literatura gris y estudios en curso.

Criterios de selección

Ensayos controlados aleatorios que compararan el agregado de vitamina K versus placebo en pacientes que inician el tratamiento con warfarina o que ya reciben tratamiento con warfarina.

Obtención y análisis de los datos

Dos autores de la revisión, de forma independiente, seleccionaron y extrajeron los datos de los estudios incluidos. Cuando hubo disconformidad, un tercer autor ayudó a lograr un consenso. También se evaluó el riesgo de sesgo.

Resultados principales

Se identificaron dos estudios con un total de 100 participantes para su inclusión en la revisión. Se encontró que el riesgo general de sesgo fue incierto en varios dominios. Ningún estudio informó el tiempo hasta la primera RIN en rango. Sólo un estudio (70 participantes) informó el tiempo medio en el rango terapéutico como porcentaje. Este estudio halló que en el grupo de participantes que presentaba un control deficiente de la RIN, el agregado de 150 microgramos (mcg) de vitamina K oral mejoró significativamente el control de la anticoagulación en los pacientes con inestabilidad no explicada de la respuesta a la warfarina. El segundo estudio (30 participantes) informó el efecto de 175 mcg de vitamina K oral versus placebo en los participantes con variabilidad alta en los niveles de la RIN. El estudio estableció la conclusión de que la administración de suplementos de vitamina K no mejoró significativamente la estabilidad de la anticoagulación para los participantes que recibían tratamiento para la anticoagulación crónica. Sin embargo, el estudio sólo estuvo disponible en forma de resumen, y la comunicación con el autor principal confirmó que no había ninguna publicación adicional. Por lo tanto, se interpretó esta conclusión con cautela. Ningún estudio informó eventos tromboembólicos, hemorragia o muerte causados por el agregado de suplementos de vitamina K.

Conclusiones de los autores

Dos estudios incluidos en esta revisión compararon si el agregado de una dosis baja (150 a 175 mcg) de vitamina K administrada a los participantes con una respuesta de variabilidad alta a la warfarina mejoraba el control de la RIN. Un estudio demostró una mejoría significativa, mientras que otro estudio más pequeño (publicado en forma de resumen solamente) no indicó ningún beneficio general. Actualmente, no hay datos suficientes para sugerir un beneficio general. Se necesitan ensayos más amplios y de calidad más alta para examinar si la vitamina K en dosis baja mejora el control de la RIN en los que inician o ya reciben tratamiento con warfarina.

Plain language summary

The addition of vitamin K to improve anticoagulation stability for patients starting or already on warfarin

People with irregularity in heart activity, mechanical heart valves, and clotting disorders are at increased risk of developing blood clots, which could lead to stroke or death. Taking warfarin significantly reduces this risk. However, taking too much warfarin can lead to excessive bleeding, while taking too little reduces its benefit. To monitor this, patients taking warfarin must have regular blood tests to check if their dose of warfarin is stable enough to find the correct balance. There is some evidence that adding a small dose of vitamin K to the warfarin improves this balance. In this review, our primary outcomes were to assess if the addition of low-dose vitamin K to warfarin had an effect on the time taken to the first INR in range; the mean within the therapeutic range; or any adverse events, such as thromboembolic events, haemorrhage, or mortality. We found two studies that met our inclusion criteria. Neither study reported the time taken to the first INR in range. One study was only available in an abbreviated format, so we were unable to interpret the results fully. Nonetheless, it was suggested that the addition of vitamin K had no benefit. A second six-month study gave a small dose of vitamin K (150 mcg daily) or placebo to participants taking warfarin with existing poor INR control. This study reported the mean time in therapeutic range as a percentage and found that in the group of participants deemed to have poor INR control, the addition of 150 mcg oral vitamin K significantly improved their anticoagulation control. However, the study was relatively small. Neither study reported any adverse events, such as thromboembolism, haemorrhage, or death. We conclude that further larger, higher quality studies are needed to conclude whether adding vitamin K to warfarin for patients starting or already on warfarin improves their anticoagulation control.

Résumé simplifié

Ajout de vitamine K pour améliorer la stabilité de l'anticoagulation chez les patients commençant ou recevant déjà de la warfarine

Les personnes atteintes d'irrégularité au niveau de l'activité cardiaque, des valves cardiaques mécaniques et de la coagulation sont à risque accru de développer des caillots de sang, ce qui pourrait entraîner un accident vasculaire cérébral (AVC) ou un décès. L'administration de la warfarine réduit significativement ce risque. Cependant, une dose trop élevée de warfarine peut entraîner des saignements excessifs, tandis qu'une dose trop faible réduit ses effets bénéfiques. Pour surveiller cela, les patients sous warfarine doivent effectuer des analyses sanguines régulières pour vérifier si leur dose de warfarine est suffisamment stable afin de trouver le juste équilibre. Certaines preuves indiquent que l'ajout d'une faible dose de vitamine K à la warfarine améliore cet équilibre. Dans cette revue, nos critères de jugement principaux étaient d'évaluer si l'ajout d'une faible dose de vitamine K à la warfarine avait un effet sur le temps nécessaire au premier RIN en marge, sur la moyenne de la marge thérapeutique, ou si elle provoquait des effets indésirables, tels que des événements thromboemboliques, des hémorragies ou une mortalité. Nous avons identifié deux études qui remplissaient nos critères d'inclusion. Aucune étude ne rapportait le temps nécessaire au premier RIN en marge. Une étude était uniquement disponible en format abrégé, nous n'étions donc pas en mesure d'interpréter les résultats de manière exhaustive. Néanmoins, il a été suggéré que l'ajout de vitamine K ne présentait aucun bénéfice. Une deuxième étude de six mois administrait une faible dose de vitamine K (150 mcg par jour) ou un placebo aux participants sous warfarine avec un contrôle existant du RIN de mauvaise qualité. Cette étude a rapporté la durée moyenne de la marge thérapeutique en pourcentage et a constaté que dans le groupe de participants jugés avoir un contrôle inadéquat du RIN, l'ajout de 150 mcg de vitamine K par voie orale améliorait significativement le contrôle de l'anticoagulation. Cependant, l'étude était de taille relativement petite. Aucune étude ne rapportait d'effets indésirables, tels que la thromboembolie, l'hémorragie ou les décès. Des études supplémentaires de plus grande taille et de meilleure qualité sont donc nécessaires pour déterminer si l'ajout de vitamine K chez les patients commençant ou recevant déjà de la warfarine améliore le contrôle de leur anticoagulation.

Notes de traduction

Translated by: French Cochrane Centre

Translation supported by: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé

平易な要約

ワルファリンを開始するまたは既に投与中の患者に対する抗凝固管理改善のためのビタミンK追加

不規則な心臓の働き、機械弁、凝固障害を有する患者は血栓発生のリスクが高まり、脳卒中や死亡につながり得る。ワルファリン投与でこのリスクは大きく低下する。しかし、ワルファリンの投与量が多すぎると過度の出血につながり、少なすぎるとその利益が低下し得る。これをモニターするには、ワルファリン投与中の患者は、定期的に血液検査を受け、自身のワルファリン用量が十分安定していて正しいバランスを見つけられるかを確認しなくてはならない。少量のビタミンKをワルファリンに加えることがこのバランスを改善するというある程度のエビデンスがある。本レビューでは、主要アウトカムは、低用量ビタミンKのワルファリンへの追加が初回INR範囲内までの時間、治療域内の平均値、血栓塞栓イベントなどの何らかの有害事象、出血、死亡率に影響するかどうかを評価することである。試験2件が選択基準を満たしたことがわかった。いずれの試験も初回INR範囲内までの時間を報告していなかった。 1件の試験は短縮形式のみで利用したため、結果を完全に解釈することはできなかった。しかし、ビタミンK追加には利益がないことが示唆されていた。第二の6カ月間の試験では、ビタミンK低用量(150 mcg/日)またはプラセボを、既存のINRコントロール不良があるワルファリン投与中の患者に投与した。この試験では、治療域内の平均時間をパーセンテージで報告し、INRコントロールが不良と考えられる群の患者においては、経口ビタミンK 150 mcg/日追加により抗凝固コントロールが有意に改善したことがわかった。しかし、この試験は比較的小規模であった。いずれの試験も血栓塞栓症、出欠、死亡などの有害事象を報告していなかった。ビタミンK追加が、ワルファリンを開始するあるいは投与中の患者にとって抗凝固コントロールを改善するためには、より大規模な質の高いさらに試験が必要である。

訳注

《実施組織》厚生労働省「「統合医療」に係る情報発信等推進事業」(eJIM:http://www.ejim.ncgg.go.jp/)[2016.1.9]
《注意》この日本語訳は、臨床医、疫学研究者などによる翻訳のチェックを受けて公開していますが、訳語の間違いなどお気づきの点がございましたら、eJIM事務局までご連絡ください。なお、2013年6月からコクラン・ライブラリーのNew review, Updated reviewとも日単位で更新されています。eJIMでは最新版の日本語訳を掲載するよう努めておりますが、タイム・ラグが生じている場合もあります。ご利用に際しては、最新版(英語版)の内容をご確認ください。

Resumen en términos sencillos

Agregado de vitamina K para mejorar la estabilidad de la anticoagulación en pacientes que inician o que ya reciben tratamiento con warfarina

Los pacientes con irregularidades en la actividad cardíaca, válvulas cardíacas mecánicas y trastornos de coagulación presentan un mayor riesgo de desarrollar coágulos sanguíneos, que podrían dar lugar a un accidente cerebrovascular o la muerte. La administración de warfarina reduce significativamente dicho riesgo. Sin embargo, la administración de demasiada warfarina puede dar lugar a hemorragia excesiva, mientras que si se administra muy poca, se reduce su beneficio. Para controlar lo anterior, los pacientes que reciben warfarina deben ser sometidos a análisis de sangre regulares para verificar si la dosis de la warfarina es lo bastante estable como para encontrar el equilibrio correcto. Hay algunas pruebas de que el agregado de una dosis pequeña de vitamina K a la warfarina mejora este equilibrio. En esta revisión, los resultados primarios fueron evaluar si el agregado de vitamina K en dosis baja a la warfarina tenía un efecto sobre el tiempo hasta la primera RIN en rango; la media dentro del rango terapéutico; o cualquier evento adverso, como los eventos tromboembólicos, la hemorragia, o la mortalidad. Se encontraron dos estudios que cumplieron los criterios de inclusión. Ningún estudio informó el tiempo hasta la primera RIN en rango. Un estudio sólo estuvo disponible en un formato abreviado, de manera que no fue posible interpretar los resultados plenamente. No obstante, se sugirió que el agregado de vitamina K no presentó ningún beneficio. Un segundo estudio de seis meses administró una dosis pequeña de vitamina K (150 mcg diarios) o placebo a los participantes que recibían warfarina y que presentaban un control deficiente de la RIN. Este estudio informó el tiempo medio en rango terapéutico como un porcentaje y halló que en el grupo de participantes con un control deficiente de la RIN, el agregado de 150 mcg de vitamina K oral mejoró significativamente el control de la anticoagulación. Sin embargo, el estudio fue relativamente pequeño. Ningún estudio informó eventos adversos, como tromboembolia, hemorragia, o la muerte. Se establece la conclusión de que se necesitan estudios adicionales más amplios y de calidad más alta para conocer si el agregado de vitamina K a la warfarina para los pacientes que inician o ya reciben tratamiento con warfarina mejora el control de la anticoagulación.

Notas de traducción

La traducción y edición de las revisiones Cochrane han sido realizadas bajo la responsabilidad del Centro Cochrane Iberoamericano, gracias a la suscripción efectuada por el Ministerio de Sanidad, Servicios Sociales e Igualdad del Gobierno español. Si detecta algún problema con la traducción, por favor, contacte con Infoglobal Suport, cochrane@infoglobal-suport.com.

Laički sažetak

Dodatak vitamina K za poboljšanje antikoagulacijske stabilnosti za pacijente koji započinju s terapijom varfarinom ili su već na terapiji varfarinom

Osobe s nepravilnostima u srčanoj aktivnosti, mehaničkim srčanim zaliscima i poremećajima zgrušavanja krvi su pod povećanim rizikom od razvoja krvnih ugrušaka koji mogu dovesti do moždanog udara ili smrti. Uzimanje varfarina značajno smanjuje taj rizik. Međutim, uzimanje previše varfarina može dovesti do velikog krvarenja, dok uzimanje premalo varfarina smanjuje njegovu učinkovitost. Za praćenje te terapije, pacijenti koji uzimaju varfarin moraju provoditi redovite krvne testove kako bi se provjerilo je li njihova doza varfarina dovoljno stabilna da se može pronaći ispravna ravnoteža. Postoje dokazi da se ta ravnoteža poboljšava dodatkom male doze vitamina K varfarinu. U ovom Cochraneovom sustavnom pregledu literature, naši primarni ishodi bili su procijeniti da li dodatak niske doze vitamina K varfarinu ima utjecaj na vrijeme potrebno za postizanje prvog INR-a u terapijskom rasponu; prosjek unutar terapijskog raspona; ili bilo koju nuspojavu kao što su tromboembolijski događaji, krvarenje ili smrtnost. INR je kratica za internacionalni normirajući omjer (engl. international normalized ratio), laboratorijski tekst kojim se prati vrijeme potrebno za zgrušavanje. Pronašli smo dva istraživanja koja su zadovoljila naše kriterije za uključivanje. Niti jedno istraživanje nije prikazalo vrijeme potrebno za postizanje prvog INR-a u terapijskom rasponu. Jedno istraživanje je bilo dostupno samo u skraćenom obliku, tako da nismo bili u mogućnosti u cijelosti procijeniti rezultate. Ipak, to istraživanje ukazuje kako dodatak vitamina K nije bio koristan. U drugom istraživanju, koje je trajalo šest mjeseci, je pacijentima na terapiji varfarinom sa slabom kontrolom INR-a, davana ili mala doza vitamina K (150 mikrograma dnevno) ili placebo. U tom istraživanju je prosječno vrijeme u terapijskom rasponu izraženo kao postotak i nađeno je da je u grupi sudionika sa slabom kontrolom INR-a dodatak od 150 mikrograma oralnog vitamina K značajno poboljšao antikoagulacijsku kontrolu. Međutim, to istraživanje je bilo relativno malo. Niti u jednom istraživanju nisu zabilježeni neželjeni događaji kao što su tromboembolizam, krvarenje ili smrt. Zaključili smo da su potrebna dodatna veća kvalitetnija istraživanja da bi se moglo zaključiti da li dodavanje vitamina K varfarinu kod pacijenata koji započinju ili su već na tretmanu varfarinom poboljšava njihovu antikoagulacijsku kontrolu.

Bilješke prijevoda

Hrvatski Cochrane
Prevela: Aida Kulo Ćesić
Ovaj sažetak preveden je u okviru volonterskog projekta prevođenja Cochrane sažetaka. Uključite se u projekt i pomozite nam u prevođenju brojnih preostalih Cochrane sažetaka koji su još uvijek dostupni samo na engleskom jeziku. Kontakt: cochrane_croatia@mefst.hr

Background

Description of the condition

A substantial number of people require oral anticoagulants: In the United Kingdom (UK), for example, 1.4% of the population require long-term treatment with anticoagulants (NICE 2010). There has been a substantial increase in the use of oral anticoagulants, particularly in the ageing population (van Walraven 2009). The numbers are going to continue to increase by about 10% each year, primarily driven by their use in people with atrial fibrillation (AF) (DTB 2009). Other reasons for the increase include improvements in clinical outcomes (Manotti 2001) and improvements in anticoagulant safety (Ansell 2001).

There are numerous other medical conditions apart from AF that lead to thromboembolic events, including deep vein thrombosis (DVT), cardiovascular causes like mechanical heart valve replacement, cardioversion, cardiomyopathy, and antiphospholipid syndrome (Baglin 2006). Oral anticoagulation therapy with vitamin K antagonists (VKAs) reduces these events (Connolly 1991; Ezekowitz 1992; Go 2003). The main oral VKA used in the UK is warfarin (McIlroy 2009). Vitamin K is a cofactor needed for the liver synthesis of factors II, VII, IX, and X, all of which are involved in the coagulation cascade. VKAs inhibit the regeneration of vitamin K hydroquinone from vitamin K epoxide by inhibiting the reductase enzymes in the vitamin K cycle (Choonara 1988). The duration of the use of warfarin depends on the medical conditions; for an isolated calf vein thrombosis, warfarin is indicated for six weeks, whereas for mechanical heart valve, it is indicated for lifelong use (Baglin 2006).

Whilst warfarin prevents thromboembolic events, it can also cause adverse effects like major haemorrhage, particularly if too much warfarin is being taken and over-anticoagulation occurs (Wan 2008). Similarly, when not enough warfarin is administered, under-anticoagulation may lead to thrombosis, hence, worsening the medical condition the clinician is trying to treat (Wan 2008). The Medicines and Healthcare Products Regulatory Agency (MHRA) received 2233 adverse reaction reports associated with warfarin use between June 1963 and June 2008. The majority of these adverse reactions reported were as a result of over-anticoagulation and bleeding, and the majority of the fatal cases reported were again associated with haemorrhage (208 of the 297 were fatal reports) (MHRA 2009). This was supported by the Adverse Effect Event Monitoring system in the United States (US). From 1993 to 2006, warfarin caused 9766 bleeding cases, including 8415 (86%) cases that led to serious complications including death, hospitalisation, or required intervention. It also showed the reporting of 635 cases as under-anticoagulation with warfarin and 511 cases having problems with coagulopathy (Wysowski 2007).

Therefore, it is important to monitor warfarin levels regularly to prevent under- or over-anticoagulation. Current models of oral anticoagulation management within the UK include the traditional hospital outpatient model and various forms of community-based models, all requiring patient attendance at a clinic (Fitzmaurice 2002). In other countries, such as Canada, a primary care physician manages oral anticoagulation (Sunderji 2004).

The international normalised ratio (INR) is used to monitor the therapeutic level of warfarin. This level varies according to the condition being treated. For example, an INR of 2.0 to 3.0 is adequate for the prevention of thromboembolic events in AF, and an INR of 3.0 to 4.0 is adequate for mitral valve replacement. So, for the former example, if INR is less than two, then under-coagulation results, and if INR is greater than three, over-anticoagulation occurs. Maintaining individuals within the narrow therapeutic ranges for INR can prove challenging in routine clinical practice. The percentage of time of INR in therapeutic range could be as low as 29% (Wan 2008). A study of longitudinal INR levels among a cohort of participants with AF showed only 33% of participants' INR was in therapeutic range (Rosenman 2009).

Description of the intervention

At the initiation stage of warfarin treatment, it is important to select the correct warfarin dose and to maintain individuals within their therapeutic range (Heneghan 2010). Attempts to stabilise patients on warfarin include checking daily INR levels after the introduction of the initial dose until INR results are in the therapeutic INR range. If one of the known medications that interact with warfarin is initiated or an intercurrent illness coexists, there will be an increase in the frequency of INR monitoring, and warfarin doses will be readjusted and monitored again (Ford 2008). Another attempt at stabilisation is to self-test the INR levels and to self-monitor the warfarin dose with the help of computer programs by patients themselves (Heneghan 2006).

Despite these attempts, however, it has been shown that up to half of all patients who receive warfarin to control coagulation fail to stabilise within their target range, particularly in the first five days of treatment (Heneghan 2006). The pharmacokinetics of warfarin can be affected by various dietary substances that contain vitamin K and also medications which then result in under- or over-anticoagulation (Holbrook 2005). Interactions with certain medications, such as macrolides, antibiotics, nonsteroidal anti-inflammatory drugs, lipid-lowering agents, and amiodarone; and certain foods that are rich in vitamin K, such as brussel sprouts and broccoli, can interfere with the warfarin blood levels, and over- or under-anticoagulation subsequently occurs (Holbrook 2005).

There is little storage of vitamin K in the body; therefore, the production of vitamin K-dependent clotting factors and proteins are highly dependent on dietary vitamin K. A brief period of reduced intake of vitamin K can cause warfarin sensitivity, and an increased intake of vitamin K-containing foods can reduce anticoagulation; both of these effects can last afterwards for several days (Franco 2004). Patients with unstable control of anticoagulation have a consistently and significantly lower intake of vitamin K than their stable counterparts matched for age, sex, and indication for warfarin (Sconce 2005). Patients who were allocated to an 80% decrease of vitamin K intake increased their INR by almost 30% seven days after the intervention. Similarly, it was estimated by the dietary records that each increase in 100 micrograms (mcg) of vitamin K intake reduced the INR by 0.2 (Rohde 2007).

In order to have a relatively stable control of warfarin in practice, clinicians often advise patients to eat a relatively similar amount of vitamin K-containing foods on a regular basis rather than eating a large serving occasionally. There are sufficient data to suggest that a constant dietary intake of vitamin K that meets current dietary recommendations of 65 to 80 micrograms/day is the most acceptable dietary advice that is given to patients on warfarin therapy (Booth 1999).

This approach is supported by randomised controlled trial evidence, which shows that 74% of participants mainly with mechanical heart valves or AF on dietary vitamin K-guided management strategy were on target for the prespecified INR at 90 days compared with 58% of participants managed conventionally (de Assis 2009). However, to have a stable intake of vitamin K-containing foods in practice is often difficult if not impossible to achieve. 

How the intervention might work

Having a regular oral vitamin K supplement taken together with a maintenance dose of warfarin improves the stability of INR (Rombouts 2007; Sconce 2007). In one randomised controlled trial, vitamin K supplementation resulted in a significant decrease in the standard deviation of INR compared with placebo, as well as a significantly greater increase in percentage time within target INR range in those patients with unstable INR despite being on warfarin for nine months (Sconce 2007).

Another prospective, randomised, placebo-controlled trial showed that the number of participants in therapeutic range for the duration of the trial doubled in the vitamin K supplementation group compared to the placebo group (43% versus 24%) (Rombouts 2007).

The INR should be checked within a few days of vitamin K initiation to allow for titration of the warfarin dose, and in order to maintain the INR in the therapeutic range, it should be monitored closely in the weeks thereafter (Ford 2007). However, as vitamin K is taken alongside warfarin and the monitoring schedule does not change greatly (and may even reduce the frequency of monitoring), the assumption is that there will be high acceptability of the intervention.

Why it is important to do this review

It is challenging to achieve and maintain the INR within the therapeutic range without concomitant increases in adverse events. Some medical conditions like DVT require the stability of INR to be achieved as quickly as possible and to continue to maintain in the therapeutic range to reduce adverse events, which in turn will reduce concomitant treatments, such as heparin, or hospital admissions, and reduce costs (Heneghan 2010).

Improving INR control is beneficial in stroke prevention, but it has also been shown to be cost-effective. An analysis of a 1000-patient cohort (mean age 70 years, atrial fibrillation at moderate-to-high risk of stroke, lifetime analysis) showed that the total number of primary and recurrent ischemic strokes was 984 with real-world INR control at a cost of USD 84,518 per patient (Sorensen 2009). However, if such patients had INR values that were always within target range, this would drop to 626 with a cost per patient of USD 68,039 (Sorensen 2009).

An economic model analysed the cost of suboptimal oral anticoagulation and showed the following: If 50% of those not receiving warfarin prophylaxis had optimal anticoagulation, 19,380 emboli would be prevented, and 1.1 billion US dollars could be saved. If 50% of those currently receiving warfarin as part of routine medical care had optimal anticoagulation, 9852 emboli would be prevented, and 1.3 billion US dollars could be saved (Caro 2004).

One possible way to achieve effective anticoagulation control is to have concomitant oral vitamin K supplement. The cost of one tablet of 1 mg of vitamin K is £0.34 (BNF 2013), which would need to be taken into account when considering the wider role-out of low-dose vitamin K to all patients on warfarin. However, low-dose vitamin K could represent a relatively low-cost method for improving INR control for patients taking warfarin.

A systematic review is needed to look for evidence that may have substantial implications in clinical and financial terms, as there has not been any review on this subject.

Objectives

To assess the effects of concomitant supplementation of low-dose oral vitamin K for anticoagulation control in patients being initiated on or taking a maintenance dose of warfarin.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCTs) of participants on warfarin in primary care or hospital settings taking concomitant oral vitamin K, compared with placebo or no treatment.

Types of participants

Adults who are over 18 years old and are on warfarin irrespective of the indication for treatment (e.g. valve replacement, AF).

Types of interventions

We included two types of intervention in this review.

Intervention 1: Oral vitamin K (of various doses), which has been added to the loading dose of warfarin. These participants were followed for the duration of the intervention, which is the first five days after initiation of warfarin.

Intervention 2: Oral vitamin K (of various doses), which has been added to the maintenance dose of warfarin. This is a lifelong intervention.

These interventions were compared to either control or placebo groups.

Types of outcome measures

Primary outcomes
  • Time taken to the first INR in range (intervention one).

  • Mean time of therapeutic range (TTRs) (intervention one).

  • Mean time in therapeutic range (intervention two).

  • Thromboembolic events included stroke, arterial embolism, symptomatic DVT, or pulmonary embolism (both interventions).

  • Major haemorrhage (includes (a) fatal bleeding; (b) symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intra-articular, or pericardial bleeding, or intramuscular with compartment syndrome; (c) bleeding causing a fall in haemoglobin level of 20 g/L-¹ (1.24 mmol-¹) or more, or leading to transfusion of two units of packed red blood cells, or a combination of the aforementioned. (both interventions).

  • Mortality (both interventions).

Secondary outcomes
  • Proportion of supratherapeutic INRs and subtherapeutic INRs (both interventions).

  • Minor haemorrhage (all other haemorragic conditions that are not included in the criteria for major haemorrhage as stated in the primary outcome) (both interventions).

  • Rescue medication needed (for example, additional dose of vitamin K) (both interventions).

  • Cost-effectiveness (both interventions).

  • Quality of life (both interventions).

Search methods for identification of studies

Electronic searches

Search strategies were developed through an iterative process combining subject headings and free-text terms for our population and intervention (Appendix 1). Methodological search filters have been used where appropriate to restrict the search to randomised controlled trials; a Cochrane sensitivity-maximising RCT filter (Lefebvre 2011) has been applied in MEDLINE. No date or language restrictions were applied. In March 2011, we searched for previously published reviews on the Database of Abstracts of Reviews of Effects (DARE in The Cochrane Library, Wiley) (Issue 2, 2011). We identified primary studies by searching the following bibliographic databases:

  • the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library, Wiley) (Issue 2, 2014);

  • Ovid MEDLINE (R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE (R) (OvidSP) (1946 to 25 February 2014);

  • Embase (OvidSP) (1974 to week 8 of 2014);

  • NHS Economics Evaluations Database (NHS EED) (The Cochrane Library, Wiley) (Issue 2, 2014); and

  • Science Citation Index Expanded™ (SCI™ Expanded) & Conference Proceedings Citation Index - Science (CPCI-S) (Web of Science™) (1945 to 27 February 2014).

The CENTRAL, MEDLINE, Embase, & NHS EED searches were conducted up to 25 February 2014; the SCI™ Expanded and CPCI-S searches were conducted up to 27 February 2014.

Searching other resources

We performed citation searches and reviewed the references of all full-text papers retrieved. We contacted experts in the field where relevant. We identified ongoing trials that were registered with the WHO International Clinical Trials Registry Platform (ICTRP) (http://apps.who.int/trialsearch/), Clinicaltrials.gov (clinicaltrials.gov), and the Current Controlled Trials Register (www.controlled-trials.com) up to 25 February 2014. We identified additional grey literature through a search of OpenGrey (www.opengrey.eu/) up to 25th February 2014. We contacted authors of included studies for additional data and information on ongoing and unpublished trials.

Data collection and analysis

Selection of studies

Two authors (KM and DN) independently assessed relevant titles. We initially excluded irrelevant studies based on title alone and excluded further studies after reviewing the title and abstract. We obtained the full texts of articles deemed to be potentially suitable for inclusion and assessed them against inclusion criteria. At each stage, the two authors (KM and DN) resolved all disagreements through discussion with a third author (CH) until they reached a consensus agreement.

Data extraction and management

We collected data on participants, interventions, and outcomes using a specifically designed data extraction form. Two authors (KM and DN) carried out independent dual extraction of data; they resolved differences between their extraction by discussion and, where necessary, in consultation with a third author (CH). In cases where the data were insufficiently reported, we made attempts to contact the relevant authors.

Assessment of risk of bias in included studies

Two review authors (KM and DN) independently assessed risk of bias using The Cochrane Collaboration's tool for assessing risk of bias. The specific aspects assessed included method of randomisation, allocation concealment, blinding of outcome assessors, treatment of incomplete outcome data, selective reporting, and other potential sources of bias. We referred to the Cochrane Handbook for Systematic Reviews of Interventions for guidance (Higgins 2011).

Measures of treatment effect

For dichotomous outcomes, we had aimed to compare different regimens using relative risks (RR) and calculate 95% confidence intervals (CIs). For continuous outcomes, we had aimed to use weighted mean difference (WMD) with 95% CIs to summarise the pooled effect. However, there were insufficient data to undertake meta-analysis in this review, but we plan to do this in future updates.

Unit of analysis issues

Where there were studies that used different ways to present the data (for example, with regard to the maintenance of warfarin, the unit of analysis issues may be days in therapeutic range or proportion of participants in therapeutic range), we had planned to contact the authors for clarification by requesting their raw data. We aimed to dichotomise the data so that it was suitable for meta-analysis where possible. We referred to the Cochrane Handbook for Systematic Reviews of Interventions for guidance (Higgins 2011). However, there were insufficient data to undertake meta-analysis in this review, but we plan to do this in future updates.

Dealing with missing data

Where any missing data were present, we contacted the study authors to obtain further relevant details. We only analysed data that were made available to us and had planned to discuss the impact of the missing data in our findings.

Assessment of heterogeneity

We had aimed to use the I² statistic to quantify the level of statistical heterogeneity (Higgins 2011). Where no heterogeneity was present, we had aimed to perform a fixed-effect meta-analysis. Where substantial heterogeneity (I² statistic above 50%) was present, we had planned to consider the potential explanations for this.

Assessment of reporting biases

We had planned to generate a funnel plot to assess publication bias. However, there were insufficient trials to carry this out; we plan to do this in future updates.

Data synthesis

We intended to perform a meta-analysis for a pooled estimate. However, there were insufficient data in this review to carry this out, but we plan to do this in future updates. 

Subgroup analysis and investigation of heterogeneity

There were insufficient numbers of included studies to carry out subgroup analysis. However, we plan to do this in future updates.

Sensitivity analysis

There were insufficient numbers of included studies to carry out sensitivity analysis. However, we plan to do this in future updates.

Results

Description of studies

See the 'Characteristics of included studies' tables and the 'Characteristics of excluded studies' tables.

Results of the search

We obtained a total of 4031 references after executing the search strategy. Of these, we deemed 98 papers to be of potential inclusion based on title alone. We excluded 80 of these after reviewing the title and abstract. We examined 18 papers in full. Of these, we excluded 16 for not being relevant. We give our reasons for exclusion in the 'Characteristics of excluded studies' tables. We carried out citation searching of all excluded papers, although this did not reveal any studies for inclusion. Therefore, we included two studies in this review. Figure 1 is a flow diagram of our search results.

Figure 1.

Study flow diagram

Included studies

We included two randomised controlled trials, involving 100 participants (Dalloul 2010; Sconce 2007). The Sconce 2007 study randomised 70 participants that were already on warfarin but deemed to have poor control to six months of either 150 mcg vitamin K daily supplementation or matched placebo. The Dalloul 2010 study identified 50 patients on warfarin therapy with high variability in their INR levels, but randomised 30 participants to receive supplemented oral vitamin K (175 mcg) daily versus placebo for six months. However, the Dalloul 2010 study was an abstract only, and communication with the corresponding author confirmed that no further data or publications were available (Dalloul 2013). Please see the 'Characteristics of included studies' tables.

Excluded studies

We excluded a total of 15 studies. Two of these studies met our inclusion criteria on all other aspects except they used different alternative vitamin K antagonists to warfarin (Gebuis 2011; Rombouts 2007). The Rombouts 2007 study was a double-blind, randomised, placebo-controlled trial examining the impact of daily vitamin K supplementation on the vitamin K antagonist phenprocoumon. The Gebuis 2011 study was a double-blind, randomised, controlled trial of varying doses of vitamin K and acenocoumarol or phenprocoumon versus acenocoumarol or phenprocoumon alone. However, as our protocol planned to assess specifically the impact of vitamin K supplementation on the vitamin K antagonist warfarin, we excluded the Rombouts 2007 and Gebuis 2011 studies, but noted their results in our discussion. The Ford 2007 study was a prospective, open-label, cross-over study, so we therefore excluded it on the basis that the design did not meet our inclusion criteria. In the de Assis 2009 study, the intervention arm was a dietary vitamin K strategy, in which participants who were overcoagulated were asked to increase their consumption of three vitamin K-rich foods (lettuce, broccoli, and liver) or decrease it if they were undercoagulated. The comparator group was treated according to standard guidelines. We therefore excluded this study as there was no evidence of concomitant vitamin K supplementation when compared with a placebo arm. We excluded the Pengo 1993 study as the aim was to assess the effect of vitamin K versus warfarin discontinuation in overcoagulated participants. We excluded the Shopnick 1998 paper as it tested the effect of vitamin K in overcoagulated participants. The Kim 2001 study examined the effect of a single dose of warfarin versus a single dose of warfarin plus 10 mg of vitamin K in otherwise healthy individuals who would not otherwise require warfarin treatment. This study was a pharmacokinetic study only, and the authors were unanimous in excluding it given the research question specific to this review. The Marongiu 1992, Pedersen 1991, Udall 1968, and Sorano 1993 studies were not randomised controlled trials. Three excluded studies were reviews or commentary pieces (Ford 2008; JFP 2008; Patriquin 2011). The NTR314 2005 citation appeared to be the trial registration of the Rombouts 2007 study. We excluded the Zuchinali 2012 study as it appeared to be an analysis of a trial in which anticoagulation control was adjusted based on prospective dietary vitamin K intake. Please see the 'Characteristics of excluded studies' tables.

Risk of bias in included studies

We assessed bias using Higgins 2011 as our reference. We categorised studies in each area as being high, low, or unclear (see the 'Characteristics of included studies' tables): We presented the results of this assessment in 'Risk of bias' tables, as well as a 'Risk of bias' graph (Figure 2) and a 'Risk of bias' summary (Figure 3).

Figure 2.

'Risk of bias' graph: Review authors' judgements about each 'Risk of bias' item presented as percentages across all included studies

Figure 3.

'Risk of bias' summary: Review authors' judgements about each 'Risk of bias' item for each included study

Allocation

The Sconce 2007 study reported that participants were randomly allocated to two groups. However, the report provided no further details regarding either the method of allocation concealment or the randomisation methods used. The Dalloul 2010 study was available in abstract form only, and we were unable to obtain any further information to assess the method of allocation.

Blinding

The Sconce 2007 study reported that the randomly allocated participants were blinded. Participants in each arm of the study were provided with either vitamin K (phytomenadione) (in 20:80 ethanol–deionised water solution) or matching placebo (20:80 ethanol–deionised water solution). Both were dispensed to participants in a 200 mL dark-brown glass bottle (vitamin K is light-sensitive) with a 5 mL volume measuring cup every four weeks. Sconce 2007 also made attempts to ensure blinding of the outcome assessment by ensuring that the pharmacist altering the dose of warfarin was also blinded to the intervention. It was unclear if researchers blinded to the intervention carried out the final assessment of results. Dalloul 2010 was a randomised, double-blind, placebo-controlled study. However, it was published as an abstract only, and we were unable to gain further detailed information after communication with the author (Dalloul 2013).

Incomplete outcome data

The Sconce 2007 study enrolled 70 participants into the trial. Of these, two participants failed to complete the trial. The study provides details of the outcomes for both participants. (One withdrew because of intervening illness, and the other died before completing the study.) The authors state that neither case was related to the study but opted not to include these data in their final analysis. The Dalloul 2010 study describes 30 participants enrolled into the trial, but from the abstract only, we were unable to identify whether all completed the trial. It was unclear if the Sconce 2007 study used an intention-to-treat principle. The fact that the final analysis did not include two participants suggests not.

Selective reporting

The Sconce 2007 study outlined primary and secondary outcomes in the methods section. The results of these were presented in the results and discussion section. The Dalloul 2010 study was available in abstract form only.

Other potential sources of bias

The Dalloul 2010 study was available in abstract only, and it is unclear if the full study was not published because the results were equivocal.

Effects of interventions

Effect of vitamin K on improved INR control

Primary outcomes
Time taken to the first INR in range

Neither Sconce 2007 nor the Dalloul 2010 study reported data on the time taken for participants to reach an INR in range.

Mean time of therapeutic range (TTRs)

Neither Sconce 2007 nor the Dalloul 2010 study reported data on the time taken for participants to reach an INR in range.

Mean time in therapeutic range

In the Sconce 2007 study, the authors report the mean time in therapeutic range as a percentage. The 35 participants allocated to the vitamin K group had their INR monitored for six months during the intervention. The mean time in range (expressed as a percentage) was then compared to the value for those same participants in the six months prior to entering the trial. Likewise, a similar methodology was chosen for the 35 participants allocated to the control arm of the trial. The Sconce 2007 study authors reported that anticoagulation control was significantly improved in both cohorts in the six-month study period compared with the previous six months. However, they further report that the vitamin K supplementation resulted in a significantly greater improvement in the stability of anticoagulation. For the vitamin K group, the mean time in range (per cent) was 59 ± 20 before the study and 87 ± 14 after the intervention period (difference 28 ± 20, P < 0.01). For the placebo group, the mean time in range (per cent) was 63 ± 18 in the six months before commencement of the study and 78 ± 17 at the end of the study (difference 15 ± 20, P < 0.01). The authors report that there were no significant differences in measures of anticoagulation control in the six months prior to the study between the two participant cohorts. The Dalloul 2010 study did not report any data on time in range.

Other markers of improved control

The Sconce 2007 trial reported that the median number of warfarin dosage changes was significantly lower in the group receiving vitamin K supplementation compared to the placebo group (vitamin K group: five changes (range three to seven) six months before the study versus two changes (range zero to five) six months after the study, a difference of -3 (range zero to -5), P < 0.001; placebo group: five changes (range three to eight) six months before the study versus three changes (range one to eight) six months after the study, a difference of -2 (range -3 to three), P < 0.001). The Dalloul 2010 study reported no statistically significant difference in the mean number of dose adjustments after treatment with vitamin K (3.9 + 2.8 versus 4.0 + 2.1, P value not given). However, we were unable to carry out any meta-analysis due to the limitations of the available data.

The Sconce 2007 study recruited participants with unstable INR values. A participant was classified unstable if the standard deviation (SD) of INR values was greater than 0.5, and they had had at least three warfarin dose changes in the previous six months. The primary end point was the SD of INR values in the six-month study period compared with the same measurement in the six months immediately prior to the study. They reported that the SD of INR significantly improved in both the vitamin K and placebo groups, but the effect was greater in the vitamin K group (-0.24 ± 0.14 versus -0.11 ± 0.18, no P value was given for this comparison). Sconce 2007 also reported anticoagulation control improved in 33 of 35 participants receiving vitamin K supplementation, whereas in the control arm, only 24 of 33 participants receiving placebo demonstrated some degree of improvement. However, the authors carried out no statistical analysis on these results, but we calculated this to be a significant improvement (odds ratio 6.19, 95% confidence interval 1.22 to 31.26, P = 0.03) (see Analysis 1.1).

Adverse events

The Sconce 2007 trial did not report any thromboembolic events, major or minor haemorrhage, or use of rescue vitamin K in participants enrolled in either arm of the study. They reported that one participant withdrew because of an intervening illness, although they did not cite the illness. However, the authors state that the case was not related to the intervention, and subsequent unblinding of allocation revealed that the participant received placebo. The authors chose not to include this in the final statistical analysis. The Dalloul 2010 study did not provide any data on thromboembolic events, major haemorrhage, minor haemorrhage, or use of rescue vitamin K. This may in part be due to the fact that we were limited to assessing data provided within the abstract only.

Mortality

The Sconce 2007 trial reported that one participant died, although the authors did not cite a reason. However, they stated that the death was unrelated to the study, and subsequent unblinding revealed the participant to have received placebo. The Dalloul 2010 study did not provide any data on mortality.

Secondary outcomes
Proportion of supratherapeutic INRs and subtherapeutic INRs

Neither Sconce 2007 nor the Dalloul 2010 study provided specific data on the proportion of supratherapeutic INRs and subtherapeutic INRs.

Minor haemorrhage (conditions that are excluded from the criteria for major haemorrhage as stated in the primary outcome)

Neither Sconce 2007 nor the Dalloul 2010 study provided specific data on minor haemorrhages.

Rescue medication needed (for example, additional dose of vitamin K)

Neither Sconce 2007 nor the Dalloul 2010 study provided specific data on the need for rescue medication.

Cost-effectiveness

Neither Sconce 2007 nor the Dalloul 2010 study provided specific data on cost-effectiveness.

Quality of life

Neither Sconce 2007 nor the Dalloul 2010 study provided specific data on quality of life.

Discussion

Summary of main results

Only two studies from 2508 citations met the inclusion criteria for our analysis of the effects of concomitant supplementation of low-dose oral vitamin K for anticoagulation control in patients being initiated on or taking a maintenance dose of warfarin (Dalloul 2010; Sconce 2007).

One of these studies (Dalloul 2010) reported no statistically significant difference in the mean number of dose adjustments after concomitant supplementation with vitamin K and concluded that there was no improvement in INR stability with vitamin K supplementation. In contrast, Sconce 2007 reported that anticoagulation control was significantly improved in participants given concomitant vitamin K supplementation. In addition, the stability of anticoagulation, as measured by the mean time in therapeutic range, was significantly higher with supplementary vitamin K (28% versus 15%, P < .01), but there was no overall effect on warfarin dose changes between intervention and placebo arm. Sconce 2007 also reported anticoagulation control improved in 33 of 35 participants receiving vitamin K supplementation, whereas in the control arm, only 24 of 33 participants receiving placebo demonstrated some degree of improvement. However, no statistical analysis was carried out on these results by the authors, but we calculated this to be a significant improvement (odds ratio 6.19, 95% confidence interval 1.22 to 31.26, P = 0.03).

Overall completeness and applicability of evidence

Since we identified only one published study during the process of our review (Sconce 2007), we interpreted our results with caution. We obtained the remaining included study (Dalloul 2010) as an abstract only, and further communication with the corresponding author confirmed that no further publications arose from this (Dalloul 2013). In the Sconce 2007 study, the intervention group received a once-daily concomitant oral supplement of 150 mcg vitamin K, which the authors point out is approximately twice the recommended daily allowance (RDA). They justify this vitamin K dose by stating that it was deemed necessary to override any variability in dietary vitamin K intake without causing a statistically significant lowering of the INR (Sconce 2007). Furthermore, the Sconce 2007 study reported median dose changes six months before the intervention and six months during it for either arm as an outcome measure. We reported these data as a demonstration of another marker of improved control. However, such an outcome is potentially limited as it is not known what constitutes a minimum clinically significant difference in dose changes to impact on a relevant outcome. The intervention arm of the Dalloul 2010 study involved participants receiving a once-daily oral dose of 175 mcg of vitamin K, but the authors gave no details of why this dose was chosen.

Quality of the evidence

We were only able to fully assess the quality of evidence for the Sconce 2007 study. We found the overall risk of bias to be low or unclear. There were no areas of the Sconce 2007 study that we deemed to have a high risk of bias. We were unable to assess the risk of bias of the Dalloul 2010 study.

Potential biases in the review process

There were insufficient numbers of included studies for us to complete a funnel plot to examine the risk of publication bias. However, we noted that the Dalloul 2010 was published in abstract only, which was confirmed through correspondence with the author (Dalloul 2013). It is possible that this occurred because there was no overall benefit shown and a full publication was not put forward. This introduces the possibility of publication bias although this is purely speculative. The fact that there are only two small included studies limits our conclusions. 

Agreements and disagreements with other studies or reviews

We included the details of only one study that was published in a peer-reviewed journal in this review (Sconce 2007). However, we made note of two excluded studies during our selection process (Gebuis 2011; Rombouts 2007). Although both studies were randomised controlled trials evaluating the effect of vitamin K supplementation on vitamin K antagonists, neither antagonist was warfarin. In the Rombouts 2007 study, the authors carried out a double-blind, randomised, placebo-controlled trial on 200 participants from an anticoagulation clinic who used the vitamin K antagonist phenprocoumon. The authors concluded that supplementation of the vitamin K antagonist phenprocoumon with 100 mcg vitamin K improved stability of anticoagulant therapy. The Gebuis 2011 study randomised 400 participants initiated on the vitamin K antagonists acenocoumarol or phenprocoumon to receive placebo or 100, 150, or 200 mcg of vitamin K1 together with their treatment. The authors concluded that in participants starting vitamin K antagonists, supplementation with low-dose vitamin K1 resulted in an improvement of time that anticoagulation was within the therapeutic range. However, the authors noted that the differences between doses were small and questioned whether the improvement was likely to be of clinical relevance. They further considered whether this result would favour vitamin K supplementation had the population consisted of only participants with unstable anticoagulant control.

Authors' conclusions

Implications for practice

Based on the availability of only one full-peer reviewed trial (Sconce 2007), our results should be interpreted with caution. However,there may besome evidence that vitamin K supplementation improves the stability of the vitamin K antagonist warfarin in patients with existing INR instability. This would be consistent with other trials that examine the effect of vitamin K supplementation on the related vitamin K antagonists acenocoumarol and phenprocoumon. However, the risk of bias in several areas of the Sconce 2007 study was unclear. In addition, the study was relatively small. A further included study suggested no clear benefit of vitamin K supplementation, although this was based on data available in abstract form only (Dalloul 2010). Based on current evidence, concomitant use of vitamin K with warfarin cannot be recommended outside of the context of a trial setting.

Implications for research

Further larger, higher quality trials are required to evaluate if vitamin K supplementation improves the INR stability in patients taking warfarin. A greater benefit may be seen in those patients with existing poor INR control. Such patients with atrial fibrillation may also be suitable for newer novel direct thrombin inhibitors, such as dabigatran (NICE 2012). However, it is unclear whether these patients would benefit from vitamin K supplementation instead.

Acknowledgements

Kamal R Mahtani is the holder of a National Institute for Health Research Academic Clinical Lectureship in General Practice. The University of Oxford Department of Primary Care Health Sciences is part of the National Institute for Health Research School for Primary Care Research, which provides financial support for senior investigators who contributed to this article.

This article presents independent research part-funded by the National Institute for Health Research (NIHR). The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.

We thank Tracy Tai for her help in writing the initial protocol.

Data and analyses

Download statistical data

Comparison 1. Other markers of improved anticoagulation control
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Other markers of improved anticoagulation control168Odds Ratio (M-H, Fixed, 95% CI)6.19 [1.22, 31.26]
Analysis 1.1.

Comparison 1 Other markers of improved anticoagulation control, Outcome 1 Other markers of improved anticoagulation control.

Appendices

Appendix 1. Search strategy

Ovid MEDLINE (R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE (R) (OvidSP) (1946 to 25 February 2014)

1. Warfarin/

2. (Warfarin or Marevan or Jantoven or Coumadin or Orfarin).mp.

3. Coumarins/

4. ((Oral* adj3 (Anticoagula* or Anti-coagula*)) or coumarin*).ti,ab.

5. 1 or 2 or 3 or 4

6. Vitamin K/

7. (Vitamin K or Menadiol or Menadione or Menaquinone or Menatetrenone or Phytonadione or Methylphytyl or Naphthoquinone or Phylloquinone or Phytomenadione).mp.

8. 6 or 7

9. 5 and 8

10. randomized controlled trial.pt.

11. controlled clinical trial.pt.

12. randomized.ab.

13. placebo.ab.

14. drug therapy.fs.

15. randomly.ab.

16. trial.ab.

17. groups.ab.

18. 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17

19. exp animals/ not human/

20. 18 not 19

21. 9 and 20

Embase (OvidSP) (1974 to week 8 of 2014)

1. *Warfarin/

2. (Warfarin or Marevan or Jantoven or Coumadin or Orfarin).mp.

3. *anticoagulant agent/po [Oral Drug Administration]

4. *coumarin anticoagulant/

5. ((Oral* adj3 (Anticoagula* or Anti-coagula*)) or coumarin*).ti,ab.

6. 1 or 2 or 3 or 4 or 5

7. exp *vitamin K group/

8. (Vitamin K or Menadiol or Menadione or Menaquinone or Menatetrenone or Phytonadione or Methylphytyl or Naphthoquinone or Phylloquinone or Phytomenadione).mp.

9. 7 or 8

10. 6 and 9

11. random*.tw. or placebo*.mp. or double-blind*.mp.

12. 10 and 11

Cochrane Central Register of Controlled Trials (CENTRAL) & NHS Economics Evaluations Database (via The Cochrane Library, Wiley) (Issue 2, 2014)

#1 Warfarin or Marevan or Jantoven or Coumadin or Orfarin:ti,ab,kw (Word variations have been searched)

#2 MeSH descriptor: [Coumarins] this term only

#3 MeSH descriptor: [Anticoagulants] this term only

#4 ((Oral* near/3 (Anticoagula* or Anti-coagula*)) or coumarin*):ti,ab,kw (Word variations have been searched)

#5 antiocoagula* or anti-coagula*:ti (Word variations have been searched)

#6 ((variab* or stable* or stability or unstable* or instable* or instability or control*) near (anticoagula* or anti-coagula*)):ti,ab,kw (Word variations have been searched)

#7 #1 or #2 or #3 or #4 or #5 or #6

#8 "Vitamin K*" or Menadiol or Menadione or Menaquinone or Menatetrenone or Phytonadione or Methylphytyl or Naphthoquinone or Phylloquinone or Phytomenadione:ti,ab,kw (Word variations have been searched)

#9 #7 and #8

Science Citation Index Expanded™ & Conference Proceedings Citation Index - Science (Web of Science™) (1945 to 27 February 2014)

# 91,026

#8 AND #7

Indexes=SCI-EXPANDED, CPCI-S Timespan=1945-2014

# 82,169,877TS=(random* OR blind* OR allocat* OR assign* OR trial* OR placebo* OR crossover* OR cross-over*)
# 73,073#6 AND #5
# 620,124 TOPIC: ("Vitamin K*" or Menadiol or Menadione or Menaquinone or Menatetrenone or Phytonadione or Methylphytyl or Naphthoquinone or Phylloquinone or Phytomenadione)
# 544,648#4 OR #3 OR #2 OR #1
# 42,682TS=((((variab* or stable* or stability or unstable* or instable* or instability or control*) NEAR/5 (anticoagula* or anti-coagula*))))
# 3719 TITLE: ((antiocoagula* or anti-coagula*))
# 227,042 TOPIC: ((((Oral* NEAR/3 (Anticoagula* or Anti-coagula*)) or coumarin*)))
# 120,094 TOPIC: ((Warfarin or Marevan or Jantoven or Coumadin or Orfarin))

Search terms for trial registries & OpenGrey

“Vitamin K” or Menadiol or Menadione or Menaquinone or Menatetrenone or Phytonadione or Methylphytyl or Naphthoquinone or Phylloquinone or Phytomenadione

Contributions of authors

Kamal R Mahtani (KM) wrote the first draft of this review with contributions from Carl Heneghan (CH), Nia Roberts (NR), and David Nunan (DN). Nia Roberts (NR) was responsible for carrying out the search strategy. All authors commented and made contributions to the final submitted review.

Declarations of interest

Kamal R Mahtani: nothing to declare.
Carl J Heneghan: nothing to declare.
David Nunan: nothing to declare.
Nia W Roberts: nothing to declare.

Sources of support

Internal sources

  • No sources of support supplied

External sources

  • Kamal Mahtani (KM) is an NIHR-funded academic clinical lecturer in General Practice, UK.

Differences between protocol and review

The authorship contributing to the full review.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Dalloul 2010

MethodsProspective, randomised, double-blind, placebo-controlled study
ParticipantsOutpatient clinic records identified 50 participants on anticoagulation therapy with high variability in the INR levels. 30 were randomised into the study
InterventionsOral vitamin K (175 mcg) daily versus placebo
OutcomesINR levels were measured on a weekly basis for the first 4 weeks of the study. For a total of 6 months during follow-up visits, INR levels were subsequently measured and handled per goal-directed therapy. The primary outcome of the study was anticoagulation stability, defined as a reduction in the number of dose modifications during follow up
NotesThis was an abstract only. No further data were available after direct contact with the author
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThis was an abstract only. However, the authors described the study as being prospective, randomised, double-blind, and placebo-controlled
Allocation concealment (selection bias)Unclear riskThis was an abstract only. No further information was available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskThis was an abstract only. However, the authors described the study as being double-blind, although it was unclear who was blinded and how
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskThis was an abstract only. However, the authors described the study as being double-blind, although it was unclear who was blinded and how
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskThis was an abstract only. 30 participants enrolled in the trial, although there were insufficient data to assess attrition
Selective reporting (reporting bias)Unclear riskThis was an abstract only
Other biasUnclear riskThis was an abstract only

Sconce 2007

MethodsRandomised controlled trial
ParticipantsParticipants were recruited from the anticoagulation monitoring clinics at the Freeman Hospital and Royal Victoria Infirmary, Newcastle upon Tyne Hospitals National Health Service (NHS) Trust. Participants with atrial fibrillation anticoagulated with warfarin for thromboembolic prophylaxis who had a target international normalised ratio (INR) range of 2.0 to 3.0, had been taking warfarin for at least 9 months, and were defined as having unstable control were eligible to take part. The authors classified a participant as unstable if the SD of his/her INR values was greater than 0.5 and he/she had had at least 3 warfarin dose changes in the previous 6 months
Interventions150 mcg oral vitamin K or placebo. Vitamin K (phytomenadione) (in 20:80 ethanol–deionised water solution) and matching placebo (20:80 ethanol–deionised water solution) were prepared as an oral solution at a concentration of 30 g/mL. Both formulations were dispensed to participants in a 200 mL dark-brown glass bottle (vitamin K is light-sensitive) with a 5 mL volume measuring cup every 4 weeks
OutcomesThe primary end point of the study was the SD of INR values in the 6-month study period compared with the same measurement in the 6 months immediately prior to the study. Secondary end points were the percentage of time at which the target INR value within 0.5 U was attained in each participant determined by the method of Azar 1994, the number of warfarin dose changes, and the number of participants who achieved an improved control of anticoagulation during the study compared with in the previous 6 months. Any adverse events, including the number and type of bleeding episodes and thromboembolic episodes including stroke, were recorded
Notes

70 participants with unstable control of anticoagulation consented to take part in the study

Fasting plasma vitamin K concentrations were measured at baseline and in the 6-month study period

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskQuote: "Seventy patients were randomly allocated to 2 groups in a double-blinded fashion." No further information was provided
Allocation concealment (selection bias)Unclear riskThere was no description of the allocation concealment technique used
Blinding of participants and personnel (performance bias)
All outcomes
Low riskThe authors made attempts to ensure that the participants were blinded to the intervention. Particpants receiving the active intervention or control were given vitamin K (phytomenadione) (in 20:80 ethanol–deionised water solution) or matching placebo (20:80 ethanol–deionised water solution). Both were dispensed to participants in a 200 mL dark-brown glass bottle (vitamin K is light-sensitive) with a 5 mL volume measuring cup every 4 weeks
Blinding of outcome assessment (detection bias)
All outcomes
Low risk

Quote: "Seventy patients were randomly allocated to 2 groups in a double-blinded fashion"

Quote: "All patients routinely attended their designated anticoagulation-monitoring service for the following 6 months, where their INR was checked and warfarin dosage adjusted if necessary using the Dawn Anticoagulation computer program (4S Information Systems, Milnthorpe, United Kingdom). This was performed independently by a pharmacist, thus preserving the study blindness." An attempt to preserve blinding had been clearly made by ensuring the pharmacist adjusting the dose remained independent. However, it was unclear if the researchers performing the outcome analysis were also blinded

Incomplete outcome data (attrition bias)
All outcomes
Low riskQuote: "Seventy patients with unstable control of anticoagulation consented to take part in the study. Of these, 2 patients failed to complete it; 1 withdrew because of intervening illness, and the other died before completing the study. Neither case was related to the study. Both patients were later identified as having received placebo, and their results were not included in the final statistic analysis." Although the 2 participants that were enrolled but did not complete the study were accounted for, we were unclear why their data were not included in the final analysis. The authors did not formally state that they used an intention-to-treat principle
Selective reporting (reporting bias)Low riskPrimary and secondary outcomes outlined in the methods section are presented and discussed in the results and discussion sections
Other biasUnclear riskQuote: "Both patients were later identified as having received placebo, and their results were not included in the final statistic analysis." This implied that analysis was not intention-to-treat

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
de Assis 2009There was no evidence of concomitant vitamin K supplementation when compared with a placebo arm
Ford 2007This was a prospective, open-label, cross-over study and was therefore excluded on the basis that the design did not meet our inclusion criteria
Ford 2008This was a review
Gebuis 2011This assessed vitamin K and acenocoumarol or phenprocoumon versus acenocoumarol or phenprocoumon alone (not warfarin)
JFP 2008This was a review
Kim 2001This was a pharmacokinetic study
Marongiu 1992This was not a RCT
NTR314 2005This appeared to be the trial registration of the Rombouts 2007 study
Patriquin 2011This was a review
Pedersen 1991This did not meet the inclusion criteria
Pengo 1993The aim was to assess the effect of vitamin K versus warfarin discontinuation in overcoagulated participants
Rombouts 2007This assessed the effect of vitamin K supplementation on the vitamin K antagonist phenprocoumon (not warfarin)
Shopnick 1998This did not meet the inclusion criteria
Sorano 1993This was not a RCT
Udall 1968This was not a RCT
Zuchinali 2012This was an analysis of a trial in which the INR was adjusted according to self-reported vitamin K intake

Characteristics of ongoing studies [ordered by study ID]

NCT00794755

Trial name or titleA phase III pilot RCT (randomized, controlled trial) to assess the effectiveness of low dose vitamin K1 (200 micrograms per day) on improving anticoagulation control in unstable patients on warfarin
MethodsA double-blind, placebo-controlled, pilot RCT: phase 3
ParticipantsUnstable patients on warfarin
InterventionsLow-dose vitamin K1 (200 micrograms per day) versus placebo
Outcomes

Primary

Anticoagulation control - point estimates (and standard deviations) for the following variables: per cent time in therapeutic range, standard deviation of INRs, number of INRs outside of therapeutic range, and number of dose changes

Recruitment numbers - number of participants deemed eligible, number of participants solicited, number of participants screened, number of participants enrolled, and number of enrolled participants lost to follow-up

Secondary

Bleeding events - both major and minor as defined by the International Society on Thrombosis and Haemostasis (ISTH) criteria

Recurrent thrombosis

Starting dateNovember 2008
Contact information-
Notes

We were unable to find the published trial for this registration. Clinicaltrials.gov reports: "No study results posted on ClinicalTrials.gov for this study"

Website accessed 12th February 2014

NCT00990158

Trial name or titleA multicentre study of low dose oral vitamin K for INR control in patients receiving warfarin
MethodsThe proposed pilot study is a multicentre, placebo-controlled, randomised trial with an additional pilot mechanistic study
ParticipantsPatients receiving warfarin
InterventionsParticipants will receive a daily dose of 150 micrograms of vitamin K or a matching placebo medication for a total of 7 months
OutcomesThe primary outcome is a simple comparison of mean TTRs in the low-dose vitamin K and placebo participants
Starting dateJuly 2010
Contact information-
NotesWebsite accessed 12th February 2014

Ancillary