Different unfractionated heparin doses for preventing arterial thrombosis in children undergoing cardiac catheterization

  • Review
  • Intervention

Authors


Abstract

Background

The role of cardiac catheterization in pediatrics has progressed significantly over the last two decades, evolving from a primary diagnostic tool to a primary treatment modality in children with congenital heart disease. Vascular complications, particularly arterial thrombosis, are among the most common unwanted post-cardiac catheterization events. In 1974, unfractionated heparin proved to be superior to placebo in decreasing the incidence of arterial thrombosis in pediatric patients. However, the optimal dose of unfractionated heparin to be utilized in this setting remains a matter of controversy.

Objectives

To evaluate the use of low-dose (< 100 units/kg) versus high-dose (≥ 100 units/kg) unfractionated heparin administered as an intravenous bolus at the time of initiation of cardiac catheterization (that is, immediately after arterial puncture), with or without subsequent heparin maintenance doses, for the prevention of post-procedural arterial thrombosis in children.

Search methods

The Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator searched the Specialised Register (last searched November 2013) and CENTRAL (2013, Issue 10). The authors searched MEDLINE, EMBASE, and the Virtual Health Library. Clinical trials databases and sources of grey literature were searched. No language restrictions were applied.

Selection criteria

Randomized or quasi-randomized trials that compared low dose to high dose unfractionated heparin administered prior to cardiac catheterization were included. We selected studies conducted in children aged 0 to 18 years.

Data collection and analysis

The first screening of potentially eligible studies was conducted by one of the authors (MLA). The second screening, quality assessment and data extraction were independently conducted by two authors (MLA, LRB). Outcomes (thrombotic events, bleeding complications, other complications) were treated as dichotomous variables. The effect measures used were risk ratio (RR), risk difference (RD) and number needed to treat (NNT), with 95% confidence intervals (CI).

Main results

Two studies with a total of 492 participants were eligible for inclusion. Risk of bias was low for all domains in one of the studies and unclear for the other. One of the trials was stopped early. The quality of evidence for our key outcomes was moderate. The CI for the risk of arterial thrombotic events was compatible with benefits of either high or low unfractionated heparin dose regimens (RR low-dose versus high-dose 1.06, 95% CI 0.58 to 1.92). Only one of the studies reported the frequency of bleeding events for the cohort of patients and found no statistically significant difference in the incidence of major and minor bleeding events between arms (RR low-dose versus high-dose 1.38, 95% CI 0.46 to 4.13 for minor bleeding; RR low-dose versus high-dose 2.96, 95% CI 0.12 to 71.34 for major bleeding events). This study also reported on the incidence of deep vein thrombosis when comparing the high versus low dose of heparin and reported a non-significant difference (RR low-dose versus high-dose 0.34, 95% CI 0.01 to 8.28). The other study lacked information about bleeding. Side effects of heparin other than bleeding complications were not reported in either of the studies.

Authors' conclusions

Due to the limitations of the current evidence, small number of included studies, and lack of details reported in one study, we are unable to determine the effects of different dosing regimens of unfractionated heparin for the prevention of vascular thrombosis during cardiac catheterization in children. A further adequately powered, randomized clinical trial is needed.

Résumé scientifique

Différentes doses d'héparine non fractionnée pour prévenir la thrombose artérielle chez les enfants subissant un cathétérisme cardiaque

Contexte

Le rôle du cathétérisme cardiaque chez l'enfant a significativement progressé durant ces deux dernières décennies, évoluant d'un outil de diagnostic primaire à une modalité de traitement primaire chez les enfants souffrant de cardiopathie congénitale. Les complications vasculaires, en particulier les thromboses artérielles, font partie des événements indésirables les plus fréquents suite à un cathétérisme cardiaque. En 1974, l'héparine non fractionnée s'avérait supérieure au placebo pour réduire l'incidence de thrombose artérielle chez les enfants. Cependant, la dose optimale d'héparine non fractionnée à être utilisée dans ce contexte reste sujette à controverse.

Objectifs

Évaluer l'utilisation d'une faible dose (< 100 unités / kg) par rapport à une dose élevée (? 100 unités / kg) d'héparine non fractionnée administrée par bolus intraveineux au moment de l'introduction du cathétérisme cardiaque (c'est à dire, immédiatement après la ponction artérielle), avec ou sans dose ultérieure d'héparine, pour la prévention de thrombose artérielle post-intervention chez les enfants.

Stratégie de recherche documentaire

Le registre des essais du groupe Cochrane sur les maladies vasculaires périphériques a effectué des recherches dans le registre spécialisé (dernière recherche en novembre 2013) et CENTRAL (2013, numéro 10). Les auteurs ont effectué des recherches dans MEDLINE, EMBASE, et la Bibliothèque Virtuelle de la Santé. Les bases de données d'essais cliniques et les sources de la littérature grise ont été consultées. Aucune restriction de langue n'a été appliquée.

Critères de sélection

Les essais randomisés ou quasi-randomisés qui comparaient une faible dose à une dose élevée d'héparine non fractionnée administrée avant un cathétérisme cardiaque ont été inclus. Nous avons sélectionné les études menées chez les enfants âgés de 0 à 18 ans.

Recueil et analyse des données

Le premier examen des études potentiellement éligibles a été réalisé par un des auteurs (MLA). Le second examen, l'évaluation de la qualité et l'extraction des données ont été réalisés de manière indépendante par deux auteurs (MLA, LRB). Les critères de jugement (événements thrombotiques, complications hémorragiques, autres complications) ont été traités sous forme de variables dichotomiques. Les mesures des effets utilisées étaient le risque relatif (RR), la différence de risques (DR) et le nombre de sujets à traiter (NST), avec des intervalles de confiance (IC) à 95 %.

Résultats principaux

Deux études portant sur un total de 492 participants étaient éligibles pour l'inclusion. Le risque de biais était faible dans tous les domaines pour l'une des études et incertain pour les autres. Un des essais a été arrêté précocement. La qualité des données pour nos principaux critères de jugement était modérée. Pour le risque d'événements thrombotiques artériels, les IC étaient compatibles avec les effets bénéfiques des schémas posologiques d'héparine non fractionnée à dose faible ou élevée (RR d'une dose faible par rapport à une dose élevée de 1,06, IC à 95 % 0,58 à 1,92). Une seule de ces études rendait compte de la fréquence d'événements hémorragiques pour la cohorte de patients et n'a trouvé aucune différence statistiquement significative dans l'incidence des événements hémorragiques majeurs et mineurs entre les groupes (RR d'une dose faible par rapport à une dose élevée de 1,38, IC à 95 % 0,46 à 4,13 pour les hémorragies mineures ; RR d'une dose faible par rapport à une dose élevée de 2,96, IC à 95 % 0,12 à 71,34 pour les hémorragies majeurs). Cette étude rapportait également l'incidence de la thrombose veineuse profonde lors de la comparaison des faibles doses d'héparine par rapport aux doses d'héparine élevées et rapportait une différence non significative (RR d'une dose faible par rapport à une dose élevée de 0,34, IC à 95 % 0,01 à 8,28). L'autre étude manquait d'informations sur les hémorragies. Les effets secondaires de l'héparine, autres que les complications hémorragiques, n'étaient rapportés dans aucune des études.

Conclusions des auteurs

En raison de preuves actuelles limitées, du nombre restreint d'études incluses et du manque de détails rapportés dans une étude, nous ne sommes pas en mesure de déterminer les effets des différents schémas posologiques d'héparine non fractionnée pour la prévention de la thrombose vasculaire lors d'un cathétérisme cardiaque chez les enfants. Un autre essai clinique randomisé, avec une puissance adéquate est nécessaire.

摘要

不同劑量的傳統肝素對接受心導管檢查兒童之動脈血栓預防療效

背景

過去20年來,心導管檢查 (cardiac catheterization) 在小兒科的地位日漸重要,從先天性心臟病 (congenital heart disease) 兒童的主要診斷工具,逐漸發展為主要的治療模組。血管併發症尤其是動脈血栓 (arterial thrombosis),為最常見的心導管術後不良事件。研究者已在1974年證實,傳統肝素(unfractionated heparin) 對降低兒童患者動脈血栓發生率的效果,優於安慰劑。但對此種狀況下應採用的最適當傳統肝素劑量,目前仍具爭議性。

目的

評估無論後續是否投予維持劑量的肝素,在開始進行心導管檢查時 (亦即在動脈穿刺後立即注射),以靜脈注射低劑量 (< 100 units/kg) 或高劑量 (≥ 100 units/kg) 傳統肝素,對預防兒童發生術後動脈血栓的效果。

搜尋策略

考科藍周邊血管疾病群組試驗 (Cochrane Peripheral Vascular Diseases Group) 搜尋協調員搜尋專業註冊 (Specialised Register) (最後一次搜尋時間為2013年11月) 和CENTRAL (2013年,第10次發行)。作者搜尋MEDLINE、EMBASE和Virtual Health Library,此外也搜尋臨床試驗資料庫和灰色文獻來源,並未設定語言限制。

選擇標準

納入於心導管檢查之前投藥,並針對低劑量與高劑量傳統肝素進行比較的隨機分配或半隨機分配試驗。我們選擇以0至18歲兒童為對象的試驗。

資料收集與分析

由1位作者 (MLA) 負責第一次篩選,選擇可能符合納入條件的試驗。由2位作者 (MLA, LRB) 負責第二次篩選,獨立進行品質評估和資料萃取。將結果 (血栓事件、出血併發症、其他併發症) 視為二元性變項處理,採用的療效測量包括風險比 (risk ratio, RR)、風險差 (risk difference, RD) 和益一需治數 (number needed to treat, NTT),信賴區間 (confidence interval, CI) 為95%。

主要結果

有2篇試驗符合納入條件,包含492名受試者。其中1篇試驗的所有領域偏差風險皆偏低,另1篇則不明。有1篇試驗提前結束。本試驗重要結果的證據品質中等。高或低劑量傳統肝素治療處方的動脈血栓風險CI,與效益不相上下 (低劑量組相對於高劑量組的RR為1.06,95% CI為0.58至1.92)。只有1篇試驗提出患者世代的出血事件發生率,但並未發現重大與輕微出血事件發生率,具有顯著的組間差異 (輕微出血:低劑量組相對於高劑量組的RR為1.38,95% CI為0.46至4.13;重大出血事件:低劑量組相對於高劑量組的RR為2.96,95% CI為0.12至71.34)。這項試驗雖也提出高、低劑量肝素組的深層靜脈血栓發生率,但並未發現顯著差異 (低劑量組相對於高劑量組的RR為0.34,95% CI為0.01至8.28)。另一篇試驗則缺乏出血相關資訊。2篇試驗均未提及出血併發症以外的肝素副作用。

作者結論

基於目前證據有限,僅能納入少數試驗,而且其中有項試驗並未提及細節,因此我們無法判斷各種傳統肝素治療處方,對預防兒童於接受心導管檢查時發生血管性血栓的效果。未來必須進行考驗力適當的隨機分配臨床試驗。

譯註


翻譯者:臺北醫學大學實證醫學研究中心。
本翻譯計畫由衛生福利部補助經費,臺北醫學大學實證醫學研究中心、台灣實證醫學學會及東亞考科藍聯盟(EACA)統籌執行。

Plain language summary

Heparin regimens to prevent arterial thrombosis in children requiring heart catheterization

Children with heart defects since birth (that is, congenital heart disease) frequently undergo a particular type of invasive procedure termed cardiac catheterization, to help diagnose or correct their condition. This procedure consists of inserting a flexible plastic tube (catheter) into the blood vessels of the patient and guiding it to his or her heart, where a dye is injected before pictures of the heart are taken. Children undergoing this type of procedure may suffer unwanted complications in the arteries into which the catheter is inserted. It is estimated that approximately 1 out of 5 to 10 patients develop arterial blood clots (that is, thrombosis) and thus require additional treatment to prevent those clots from growing further.

Heparin, a medication known as a 'blood thinner' (that is, an anticoagulant drug), employed in clinical practice since 1935, has been used to decrease the number of arterial blood clots during heart catheterization. However, the best dose of heparin to be used in children during heart catheterization remains to be determined. This review included two small randomized controlled trials with a total of 492 participants. These two studies found only imprecise evidence that the risk of arterial clots was similar in both the low-dose and high-dose heparin groups. Only one study reported bleeding complications and the frequency of clots in veins, which were also not significantly different between groups. Side effects of heparin other than bleeding complications were not reported in either of the studies. However, because the two included studies were small and because of concerns about the quality of one of them due to the lack of information provided, their results are insufficient to determine the effects of different heparin doses. Further larger studies are required to answer this question.

Résumé simplifié

Schémas posologiques d'héparine pour prévenir la thrombose artérielle chez les enfants devant subir un cathétérisme cardiaque

Les enfants présentant des anomalies cardiaques depuis la naissance (c'est à dire une cardiopathie congénitale) subissent fréquemment une procédure invasive appelée cathétérisme cardiaque, afin d'aider à diagnostiquer ou à corriger leur condition. Cette procédure consiste à insérer un tube en plastique flexible (cathéter) dans les vaisseaux sanguins du patient et à le diriger vers son cœur, où un colorant est injecté avant que les images du cœur ne soient prises. Les enfants subissant ce type de procédure peuvent souffrir de complications indésirables au niveau des artères dans lesquels le cathéter est inséré. On estime qu'environ 1 sur 5 à 10 patients développent des caillots artériels (une thrombose) et nécessitent donc un traitement supplémentaire pour prévenir le développement d'autres caillots.

L'héparine, un médicament connu comme « anticoagulant », employé dans la pratique clinique depuis 1935, a été utilisé pour diminuer le nombre de caillots artériels lors d'un cathétérisme cardiaque. Cependant, la dose d'héparine la plus adéquate pour être utilisée chez l'enfant au cours d'un cathétérisme cardiaque reste à déterminer. Cette revue a inclue deux essais contrôlés randomisés de petite taille avec un total de 492 participants. Ces deux études ont uniquement prouvé de façon imprécise que le risque de caillots artériels était similaire dans le groupe d'héparine à faible dose comparé au groupe d'héparine à doses élevée. Les complications hémorragiques et la fréquence des caillots sanguins dans les veines, rapportées par une seule étude, n'étaient également pas significativement différentes entre les groupes. Les effets secondaires de l'héparine autres que les complications hémorragiques n'étaient rapportés dans aucune des études. Toutefois, étant donné que les deux études incluses étaient de petite taille et en raison d'inquiétudes relatives à la qualité de l'une d'entre elles suite au manque d'informations fournies, leurs résultats sont insuffisants pour déterminer les effets des différentes doses d'héparine. D'autres études plus étendues sont nécessaires pour répondre à cette question.

Notes de traduction

Traduit par: French Cochrane Centre 6th August, 2014
Traduction financée par: 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

淺顯易懂的口語結論

肝素治療對接受心導管檢查兒童發生動脈血栓的預防療效

出生時心臟就有缺陷 (即先天性心臟病) 的兒童,通常需要接受一種特殊的侵入性程序,稱為心導管檢查,以協助醫師診斷或治療疾病。進行心導管檢查時,醫師必須將1條具有彈性的塑膠管 (導管),插入患者血管並導引進入心臟,在拍攝心臟的影像之前先注射染劑。接受這種檢查程序的兒童,插入導管的動脈可能會出現不良的併發症。根據估計,每5至10位患者中,約有1位患者會出現動脈血塊 (亦即血栓),因而需要接受額外治療,以免血塊持續增大。

肝素是一種稱為「血液稀釋劑」的藥物 (亦即一種抗凝血劑),自1935年起就應用於臨床實務,可降低心導管檢查時動脈出現血塊的機率。不過尚未針對接受心導管檢查的兒童,確立最佳肝素劑量。本文獻回顧納入2篇小型隨機對照試驗,共包含492名受試者。這2篇試驗只發現不甚精確的證據,顯示低劑量和高劑量肝素組的動脈血塊發生風險相似。其中只有1篇試驗提到出血併發症及靜脈血栓的發生率,但並未發現顯著的組間差異。2篇試驗都未提及出血併發症以外的肝素副作用。不過由於我們所納入的2篇試驗都是小型試驗,而且其中1篇試驗的品質因缺乏部分資訊,所以具有疑慮,因此不足以根據其結果,判斷不同肝素劑量的療效。未來必須進行較大型的試驗,才能釐清這個問題。

譯註


翻譯者:臺北醫學大學實證醫學研究中心。
本翻譯計畫由衛生福利部補助經費,臺北醫學大學實證醫學研究中心、台灣實證醫學學會及東亞考科藍聯盟(EACA)統籌執行。

Summary of findings(Explanation)

Summary of findings for the main comparison. 
  1. a 95% CI by modified Wald method (Motulsky 2010)
    b quality of evidence downgraded due to imprecision of results and potential limitations in design and implementation of one of the studies
    c quality of evidence downgraded due to imprecision of results

Low unfractionated heparin (UFH) dose compared with high UFH dose for prevention of arterial thrombosis during cardiac catheterization

Patient or population: children with congenital heart disease undergoing cardiac catheterization

Settings: hospital

Intervention: low doses of UFH (< 100 units/kg)

Comparison: high doses of UFH (≥ 100 units/kg)

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
High UFH doseLow UFH dose

Arterial thrombosis

(6-48 hours after CC)

7.7 per 100 procedures 8.2 per 100 procedures
(5.3 to 12.3)a
RR 1.06 (0.58 to 1.92)492
(2)
⊕⊕⊕⊝
moderateb

Clinical assessment in one study;

Ultrasound assessment in the other study

Minor bleeding 7.4 per 100 procedures 10.1 per 100 procedures (4 to 20)a RR 1.38 (0.46 to 4.13)

137

(1)

⊕⊕⊕⊝
moderatec
Only one study reported minor bleeding
Major bleeding 0 per 100 procedures 1.5 per 100 procedures (< 0.01 to 8)a RR 2.96 (0.12 to 71.34)

137

(1)

⊕⊕⊕⊝
moderatec
Only one study reported major bleeding
Venous thrombosis 1.5 per 100 procedures 0 per 100 procedures RR 0.34 (0.01 to 8.28)

126

(1)

⊕⊕⊕⊝
moderatec
Only one study reported venous thrombosis
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
UFH: Unfractionated heparin; CI: Confidence interval; CC: Cardiac catheterization;RR: Risk Ratio
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Cardiac catheterization (CC) was introduced in pediatrics in the 1950s (Boijsen 1968; Vlad 1964). Vascular complications, particularly thrombosis of the catheterized vessel, were immediately recognized as the most common and serious complications of this technique among children (Freed 1974; Hohn 1969; Hurwitz 1977; Stanger 1974). Despite the use of systemic anticoagulation and the improved techniques, equipment, and expertise of operators, arterial vascular occlusion remains the most frequent complication of CC in pediatric patients (Cassidy 1992; Fellows 1987; Kulkarni 2006; Mehta 2008), particularly in small infants.

Overall, arterial thrombosis has been reported in 8.5 per 10,000 pediatric hospital admissions (Monagle 2008). Catheter-related peripheral arterial thrombosis comprises the vast majority of the pediatric cases, mostly of an iatrogenic nature secondary to arterial catheterization (Albisetti 2005; Balci 2008). Well-designed prospective studies have shown the incidence of Doppler detected thrombotic complications following catheterization to be 32% in infants and 7.8% in children (Kocis 1995; Kulkarni 2006).

Thrombotic complications at the puncture site and potential extension into the ileo-femoral arterial system are serious events (Monagle 2012), particularly in cases with an immediate risk of limb loss or bleeding from the use of thrombolytic or anticoagulant therapy (Albisetti 2005; Monagle 2012). Long-term consequences of vessel obstruction include impaired limb growth, reported in 8% of children evaluated five to 14 years after catheterization (Macnicol 2000; Peuster 2000; Taylor 1990), muscle wasting, claudication, and loss of arterial access for future catheterizations in up to 30% of previously catheterized children and adolescents (Celermajer 1993), which is highly relevant to children who require repeated CC (Andrew 1997; Balaguru 2003).

Description of the intervention

In the early 1970s, observational studies conducted in adult patients showed that peri-procedural unfractionated heparin injection was associated with a lower frequency of thrombotic events at the site of insertion of angiographic catheters (Eyer 1973; Walker 1973; Wallace 1972). These initial findings legitimized the current practice of systemic anticoagulation during CC, initially much feared and warned against due to the risk of serious bleeding in the case of inadvertent vascular or cardiac perforations (Braunwald 1968; Formanek 1970). Soon afterwards systemic heparinization was introduced in pediatrics and now being part of daily current practice it has been responsible for the decline in the incidence of arterial thrombosis following CC (Girod 1982).

How the intervention might work

Unfractionated heparin, from now on referred to as heparin, is the oldest and the most widely utilized anticoagulant agent. It was listed as an essential drug in 1997 by the World Health Organization (WHO 1997). Its appeal derives from its very short half-life and immediate reversibility in case of an emergency, making it suitable for patients in critical conditions requiring anticoagulation, or for patients undergoing procedures such as CC.

The anticoagulant effect of heparin is mainly exerted by catalyzing thrombin and activated factor X inactivation.

Why it is important to do this review

The role of CC in pediatrics has progressed significantly over the last two decades, evolving from a primary diagnostic tool to a primary treatment modality in children with congenital heart disease (Wyszynski 2010). Nevertheless, utilization of new technologies (that is larger sheaths and catheters) resulting in longer procedure times has contributed to a rise in the incidence of thromboembolic events (Grady 1995). Furthermore, vascular complication rates are reportedly three to six times higher in interventional procedures than in diagnostic procedures (Ino 1988), contributing to this increase.

Congenital heart disease is one of the most common genetic abnormalities in children and is diagnosed in approximately 1% of all births. Severe conditions requiring specialized cardiologic support have been estimated to occur in 2.5 to 3 per 1000 live births (Hoffman 2002). In Canada alone, approximately 2000 children are born every year with congenital heart disease, many of whom will require several CCs throughout their childhood. As more children with complex congenital heart disease survive longer and require multiple catheterizations, preventing thrombotic events is paramount since this “can ultimately determine a child's future therapeutic options” (Grady 1995).

One of the strategies to lower the frequency of thrombotic complications during CC is the use of prophylactic systemic anticoagulation, which is administered immediately after arterial puncture. On the other hand the major complication associated with heparin is bleeding. In adult patients it has been reported that the incidence of bleeding complications associated with heparin may be influenced by the dose or the intensity of anticoagulation, as measured by laboratory tests evaluating hemostasis (Levine 2001). Consequently, the use of systemic heparinization for thrombosis prevention post-CC is associated with a narrow therapeutic window, balancing protection against clots and bleeding. In spite of the widespread use of heparin during CC, the optimal dose has not yet been established (Price 2003).

Current guidelines recommend the use of a single heparin dose of 100 units/kg administered as a bolus (Monagle 2012). This recommendation is rooted in a randomized trial conducted in 1974 which compared 100 to 150 units/kg of heparin to placebo (Freed 1974), showing a lower incidence of thrombotic complications in the heparin group (8% versus 40%). Subsequent studies have continued using this relatively high dose. Bleeding complications have been described in approximately 7.5% to 8.5% of children undergoing CC with doses of heparin equal to or higher than 100 units/kg (Mehta 2008; Vitiello 1998). It is noteworthy that a 50 unit/kg initial dose has also been utilized (Kulkarni 2006). Nonetheless, little is known about bleeding rates and laboratory alterations following the administration of distinct heparin doses in children undergoing CC (Chen 2013; Kim 2010; Newall 2010). Given the high volume of procedures and their related morbidity, either related to thrombotic or to bleeding complications, it is vital to titrate the dosing of anticoagulation prophylaxis to an optimal level that guarantees both efficacy and safety.

Objectives

To evaluate the use of low-dose (< 100 units/kg) versus high-dose (≥ 100 units/kg) unfractionated heparin administered as an intravenous bolus at the time of initiation of cardiac catheterization (that is, immediately after arterial puncture), with or without subsequent heparin maintenance doses, for the prevention of post-procedural arterial thrombosis in children.

Methods

Criteria for considering studies for this review

Types of studies

We included randomized and quasi-randomized controlled trials that compared peri-procedural prophylactic regimens of unfractionated heparin administration in children undergoing CC. We excluded studies that did not match those study designs.

Types of participants

All studies conducted in children aged 0 to 18 years undergoing CC were included. Studies conducted primarily in adults and only including a few adolescents were therefore excluded. When patients older than 18 years of age were included, we considered only those studies with a median or mean age < 10 years, since these studies include a majority of pediatric patients.

Types of interventions

We included studies in which two or more different unfractionated heparin doses administered at the time of initiation of CC (immediately after arterial puncture) were compared. The studies had to involve a low unfractionated heparin dose arm, defined as < 100 units/kg, and a high unfractionated heparin dose arm, defined as ≥ 100 units/kg. The cut-off points were taken from current pediatric guidelines, which recommend the use of heparin doses of 100 units/kg administered as a bolus, compared with lower doses of heparin (50 units/kg) (Monagle 2012). Even though the initial heparin doses were the main focus of the study, the use of subsequent maintenance doses was also registered.

Types of outcome measures

Primary outcomes
Thrombotic events

Symptomatic or asymptomatic events of arterial thrombosis diagnosed during the immediate post-procedural period following CC (first 48 hours) was the primary outcome. Clinical findings (decreased pulses lasting more than six hours after the procedure) or imaging studies (doppler ultrasound or arteriography), or both, were acceptable methods for the diagnosis of arterial thrombosis. Although clinical suspicion is not considered the gold standard method for detection of thrombus, it remains a widely accepted method for assessment of obstruction of a blood vessel in this clinical setting.

Bleeding complications

1) Major bleeding complications were defined as one or more of the following (adapted from Schulman 2010 and Mitchell 2011):

  • fatal bleeding;

  • bleeding that is symptomatic and occurs in a critical area or organ (such as intracranial, intraspinal, intraocular, retroperitoneal, pericardial, in a non-operated joint, or intramuscular with compartment syndrome), assessed in consultation with the surgeon;

  • extrasurgical site bleeding causing a fall in hemoglobin level of 20 g/L (1.24 mmol/L) or more, or leading to transfusion of whole blood or red cells with a temporal association within 24 to 48 h of the bleeding;

  • surgical site bleeding that requires a second intervention (open, arthroscopic, endovascular) or a hemarthrosis of sufficient size as to interfere with rehabilitation by delaying mobilization or causing delayed wound healing, resulting in prolonged hospitalization or a deep wound infection;

  • surgical site bleeding that is unexpected and prolonged or sufficiently large to cause hemodynamic instability, as assessed by the surgeon. There should be an associated fall in the hemoglobin level of at least 20 g/L (1.24 mmol/L), or a transfusion indicated by the bleeding with a temporal association within 24 h of the bleeding.

The period for collection of these data was from the start of the procedure until five half-lives after the last dose of the unfractionated heparin (that is, approximately 2.5 to 5 hours depending on the dose).

2) Minor bleeding complications: overt or macroscopic evidence of bleeding not encompassed in the above definition.

Secondary outcomes
  • Other thrombotic events: venous thrombosis and thrombosis in implanted devices.

  • Other side effects of heparin:

    • incidence of heparin induced thrombocytopenia (HIT), as per the definition of the included studies. Ideally, the definition of HIT should combine consistent clinical findings (thrombocytopenia occurring five to 14 days after initiation of heparin exposure, or sooner in cases with prior heparin exposure, with or without thrombosis) and laboratory assessment for HIT antibodies (platelet-activating antibodies tested with functional assays, with or without additional antigen assays). These criteria have been extrapolated from those of adult patients (Warkentin 2005);

    • anaphylaxis, as per the definition of the included studies;

    • skin necrosis, as per the definition of the included studies.

Search methods for identification of studies

We conducted a systematic search for all published articles with the assistance of an experienced librarian. The search was not limited to any language.

Electronic searches

The Cochrane Peripheral Vascular Diseases Group Trials Search Co-ordinator (TSC) searched the Specialised Register (last searched November 2013) and the Cochrane Central Register of Controlled Trials (CENTRAL) (2013, Issue 10), part of The Cochrane Library (www.thecochranelibrary.com). See Appendix 1 for details of the search strategy used to search CENTRAL. The Specialised Register is maintained by the TSC and is constructed from weekly electronic searches of MEDLINE, EMBASE, CINAHL, AMED, and through handsearching relevant journals. The full list of the databases, journals and conference proceedings which have been searched, as well as the search strategies used, are described in the Specialised Register section of the Cochrane Peripheral Vascular Diseases Group module in The Cochrane Library (www.thecochranelibrary.com).

The following trial databases were searched by the TSC for details of ongoing and unpublished studies using the terms cardiac and catheter and heparin:

Authors' searches

We searched the following electronic databases:

The authors' search strategies are detailed in Appendix 2, Appendix 3 and Appendix 4.

We also searched the following databases for ongoing or unpublished trials (date of search 7 November 2013) using the search terms detailed in Appendix 5, Appendix 6, and Appendix 7:

Searching other resources

We screened the references of identified articles for additional studies, as well as conference proceedings of the major hematology, thrombosis and cardiology meetings (American Society of Hematology (2004 to 2012), International Society of Thrombosis and Haemostasis (2003 to 2011), American Heart Association (2009 to 2011)) and other sources of grey literature such as The Grey Literature Report and the OpenGrey (System for Information on Grey Literature in Europe). In addition, we searched theses on this subject in online dissertation databases (Proquest and Open Thesis databases).

The authors' search terms are detailed in Appendix 8 and Appendix 9.

Data collection and analysis

Selection of studies

One of the authors (MLA) reviewed and evaluated the titles and abstracts of the articles retrieved during the searches to identify and exclude irrelevant studies. All articles that met the eligibility criteria based on the first screening were assessed independently by two review authors (MLA, LRB). The authors discussed and resolved disagreements by consensus or by involving the third author (PSS).

Data extraction and management

We created a specific electronic data collection form, which was used for data extraction. Two authors (MLA, LRB) individually and independently extracted data from the eligible studies. The authors discussed and resolved disagreements and consulted the third author (PSS) whenever required. Study investigators were contacted in order to retrieve full information not found in the original publications.

Assessment of risk of bias in included studies

We assessed the risk of bias in the included studies following the Cochrane risk of bias tool (Higgins 2008). Two authors independently performed this evaluation (MLA, LRB), discussing any disagreements to reach a consensus.

The following domains of risk of bias were assessed and reported for individual studies (Higgins 2008).

Selection bias (random sequence generation and allocation concealment)

For each included study, we categorized the risk of selection bias as follows.

Random sequence generation
  • Low risk - adequate (any truly random process e.g., random number table; computer random number generator)

  • High risk - inadequate (any non-random process e.g., odd or even date of birth; hospital or clinic record number)

  • Unclear risk - no or unclear information provided

Allocation concealment
  • Low risk - adequate (e.g., telephone or central randomisation; consecutively numbered sealed opaque envelopes)

  • High risk - inadequate (open random allocation; unsealed or non-opaque envelopes, alternation; date of birth)

  • Unclear risk - no or unclear information provided

Performance bias

For each included study, we categorized the methods used to blind study personnel from knowledge of which intervention a participant received.

  • Low risk - adequate for personnel (a placebo that could not be distinguished from the active drug was used in the control group).

  • High risk - inadequate, personnel aware of group assignment.

  • Unclear risk - no or unclear information provided.

Detection bias

For each included study, we categorized the methods used to blind outcome assessors from knowledge of which intervention a participant received. Blinding was assessed separately for different outcomes or classes of outcomes. We categorized the methods used with regards to detection bias as follows.

  • Low risk - adequate, follow-up was performed with assessors blinded to group assignment.

  • High risk - inadequate, assessors at follow-up were aware of group assignment.

  • Unclear risk - no or unclear information provided.

Attrition bias

For each included study, and for each outcome, we described the completeness of data including attrition and exclusions from the analysis. We noted whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total number of randomized participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. We categorized the methods with respect to the risk of attrition bias as follows.

  • Low risk - adequate (< 10% missing data).

  • High risk - inadequate (> 10% missing data).

  • Unclear risk - no or unclear information provided.

Reporting bias

For each included study, we described how we investigated the risk of selective outcome reporting bias and what we found. We assessed the methods as follows.

  • Low risk - adequate (where it is clear that all of the study's pre-specified outcomes and all expected outcomes of interest to the review have been reported).

  • High risk - inadequate (where not all the study's pre-specified outcomes have been reported; one or more reported primary outcomes were not pre-specified; outcomes of interest are reported incompletely and so cannot be used; study fails to include results of a key outcome that would have been expected to have been reported).

  • Unclear risk - no or unclear information provided (the study protocol was not available).

Other bias

For each included study, we described any important concerns we had about other possible sources of bias (for example, whether there was a potential source of bias related to the specific study design or whether the trial was stopped early due to some data-dependent process). We assessed whether each study was free of other problems that could put it at risk of bias as follows.

  • Low risk - no concerns of other bias.

  • High risk - concerns about multiple looks at the data with the results made known to the investigators; difference in number of patients enrolled in abstract and final publications of the paper.

  • Unclear - concerns about potential sources of bias that could not be verified by contacting the authors.

Measures of treatment effect

We analyzed thrombotic events and complications associated with heparin as dichotomous outcomes.

Dealing with missing data

In cases in which the dropout process was considered to be independent of the observed data and the missing data, or dependent on the observed data but independent of the missing data (missing at random), we used the available information. Though the use of available data only could potentially lead to bias, this practice is commonly used in meta-analyses and provides a sensible starting point (Higgins 2008b; Higgins 2008c).

Assessment of heterogeneity

We used the Chi2 test of homogeneity, with the significance threshold set at P < 0.10, to detect statistically significant heterogeneity. In addition, we investigated the degree of heterogeneity by calculating the I2 statistic (Higgins 2008b).

Assessment of reporting biases

We did not investigate publication bias using funnel plots since the number of studies identified was small (that is, less than 10 studies).

Data synthesis

We used Review Manager software (RevMan Version 5.1) to perform data analysis. Risk ratio (RR), risk difference (RD), number needed to treat either to benefit (NNTB) or harm (NNTH) (Altman 1998) were estimated for the outcomes when appropriate; the 95% confidence interval (CI) was calculated for each estimate. We used a random-effects model based on the inverse variance approach for our analysis.

Subgroup analysis and investigation of heterogeneity

We planned to stratify the analysis according to the patients' ages at the time of the procedure (older or younger than one year), weight (above or below 10 kg), type of procedure (intervention versus diagnostic), and any other possible sources of heterogeneity (such as duration of procedure, subsequent heparin dosing, activated partial thromboplastin time (aPTT) or antifactor Xa (anti-Xa) levels achieved during the procedure). It is hypothesized that younger patients and patients with weight below 10 kg may be at higher risk of thrombotic events due to the small caliber of their vasculature in relation to the size of the instruments used for cardiac catheterization. Similarly, interventional procedures may require relatively larger instruments, thus portending a higher risk of thrombosis. Due to unavailability of data we were unable to conduct subgroup analysis in this version of the review.

Sensitivity analysis

We planned to perform a sensitivity analysis by excluding low quality studies but, given the small number of studies included, it was not possible to conduct such an analysis.

Results

Description of studies

Results of the search

See Figure 1, which shows the study selection flowchart. References were saved in an EndNote X6 library and used to identify duplicates. One author (MLA) reviewed the references against our inclusion criteria.

Figure 1.

Study flow diagram.

Included studies

One of the studies (Hanslik 2011) randomized 137 patients aged 0 to 19 years (median 5.6 years, 23% infants) to receive either a high dose of heparin (100 units/kg), followed by a continuous infusion of therapeutic doses of heparin, or a low dose of heparin consisting of a bolus of 50 units/kg at the time of vascular access, and an additional bolus of 50 units/kg every two hours during the procedure. Pre- and post-procedural assessment of arterial thrombosis was done using Doppler ultrasound (US). Additionally, clinical assessment was recorded.

The second study (Saxena 1997) included 366 patients aged 17 days to 11 years (mean 39.5 months, 24.9% infants). Patients were randomized to receive 50 units/kg or 100 units/kg of heparin as a single bolus administered immediately after arterial puncture. Assessment of arterial thrombosis was done by clinical examination only.

For further details of the included studies, please refer to Characteristics of included studies.

Excluded studies

Please refer to Characteristics of excluded studies for details.

A total of 18 studies were excluded (Bulbul 2002; Butt 1987; Chen 2013; Downs 1974; Freed 1974; Girod 1982; Grady 1995; Hollinger 2009; Kavanagh-Gray 1978; Kim 2010; Kulkarni 2006; Netz 1987; Phillips 2010; Rao 1981; Roschitz 2002; Roushdy 2010; Schroeder 2010; Zeevi 1999). Thirteen studies were excluded because their design (11 prospective studies (Chen 2013; Downs 1974; Girod 1982; Grady 1995; Kavanagh-Gray 1978; Kim 2010; Kulkarni 2006; Netz 1987; Roschitz 2002; Roushdy 2010; Zeevi 1999), one retrospective study (Phillips 2010) and one review on the topic (Hollinger 2009)). Five randomized controlled trials (RCTs) were excluded, two because they looked into interventions after CC (Rao 1981; Schroeder 2010); two because the comparisons were not clinically meaningful, that is, heparin versus normal saline (Freed 1974) and heparin in high doses only (Bulbul 2002); and one because the setting and outcome did not match those of our review (Butt 1987).

Risk of bias in included studies

Risk of bias for all domains was low in one of the studies (Hanslik 2011) and unclear for the other (Saxena 1997). Please see the risk of bias graph (Figure 2) and 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

Randomization sequences were generated by an independent statistician, using a computer, in one of the studies (Hanslik 2011). Random numbers were used in the second study (Saxena 1997).

In one of the studies (Hanslik 2011) randomization codes were allocated using sequentially numbered, opaque sealed envelopes, which were opened by an assistant in the CC laboratory. The second study (Saxena 1997) did not provide enough information in this regard. Please refer to Characteristics of included studies for details.

Blinding

Patients, physicians performing the CC, those performing the clinical outcome assessment or vascular ultrasound, and the study coordinators were blinded to treatment allocation in one of the studies (Hanslik 2011). Not enough information was available in the second study (Saxena 1997). Please refer to Characteristics of included studies for details.

Incomplete outcome data

One of the studies (Hanslik 2011) had a total attrition of 8% (11 patients). The investigators of the study provided information regarding the lack of outcome assessment in these 11 patients, which was unlikely to be related to the outcome and was balanced overall across groups. The outcome of the second study (Saxena 1997) was reported in the same number of patients that were randomized. Please refer to Characteristics of included studies for details.

Selective reporting

Whereas all pre-specified outcomes were reported in one of the trials (Hanslik 2011), not enough information was available in the second trial (Saxena 1997). Please refer to Characteristics of included studies for additional details.

Other potential sources of bias

One of the trials (Hanslik 2011) was stopped early due to a lower than expected incidence of the outcome, as shown by an interim analysis; it must be pointed out that the interim analysis was conducted without breaking the randomization code. Insufficient information was available for the second trial (Saxena 1997). Please refer to Characteristics of included studies for further details.

Effects of interventions

See: Summary of findings for the main comparison

Primary outcomes

Arterial thrombotic events

Two studies were included in the final analysis of this outcome (Hanslik 2011; Saxena 1997). No statistically significant difference in the incidence of arterial thrombosis was found between the two arms: RR low-dose versus high-dose 1.06 (95% CI 0.58 to 1.92, P = 0.86; moderate quality evidence) (Summary of findings for the main comparison); RD low-dose versus high-dose 0.005 (95% CI -0.04 to 0.05). The NNTH for the outcome arterial thrombosis was 212, with 95% CI from a NNTH of 19, via infinity, to a NNTB of 23 (Altman 1998); I2 = 0% (Figure 4).

Figure 4.

Forest plot of comparison: 1 Primary Outcome, outcome: 1.1 Arterial thrombosis.

Bleeding complications

The study conducted by Hanslik (Hanslik 2011) reported no statistically significant difference in the incidence of major and minor bleeding events between arms: RR low-dose versus high-dose 1.38 (95% CI 0.46 to 4.13; moderate quality evidence) (Summary of findings for the main comparison); RD 0.02 (95% CI -0.06 to 0.12); NNTH 36, 95% CI from a NNTH of 8, via infinity, to a NNTB of 15 (Altman 1998) for minor bleedings; RR low-dose versus high-dose 2.96 (95% CI 0.12 to 71.34; moderate quality evidence) (Summary of findings for the main comparison); RD 0.01 (95% CI -0.02 to 0.05); NNTH 69, with 95% CI from a NNTH of 23, via infinity, to a NNTB of 73 (Altman 1998) for major bleeding events.

The study conducted by Saxena (Saxena 1997) reported bleeding events only in a subset of patients receiving additional anticoagulant or thrombolytic therapy and not in all patients; their data were therefore not included in this review.

Secondary outcomes

The study conducted by Hanslik (Hanslik 2011) also reported a non-significant difference in the incidence of deep vein thrombosis when comparing the high versus low dose of heparin: RR low-dose versus high-dose 0.34 (95% CI 0.01 to 8.28; moderate quality evidence) (Summary of findings for the main comparison); RD -0.016 (95% CI -0.06 to 0.03); NNTB 64, with 95% CI from a NNTH of 68, via infinity, to a NNTB of 22 (Altman 1998). The authors did not report on additional side effects of heparin.

The study conduced by Saxena (Saxena 1997) did not determine the incidence of venous thrombotic events. The investigators did not report additional secondary side effects of heparin in either arm of their trial.

Discussion

Summary of main results

Hanslik conducted a randomized trial (Hanslik 2011) comparing high versus low doses of heparin in 137 pediatric patients. Blinded interim analysis showed lower than expected occurrence of events, which led to recalculation of the study sample size. The required number of patients was 10 times larger than the initial estimation. The trial was therefore stopped and the data were analyzed. The other included study (Saxena 1997) provided data on the incidence of arterial thrombosis in 366 children receiving a low versus a high dose of heparin. Neither trial found differences in the incidence of arterial thrombotic events when comparing the two heparin regimens. The confidence interval for the risk of arterial thrombotic events was compatible with benefits of either high- or low-dose unfractionated heparin regimens (RR low-dose versus high-dose 1.06, 95% CI 0.58 to 1.92).

The first study (Hanslik 2011) did not find a statistically significant difference in the incidence of major or minor bleeding events between high and low heparin doses. The second study (Saxena 1997) did not provide complete information in this regard. Further details can be found in the Summary of findings for the main comparison.

Overall completeness and applicability of evidence

Two trials met our inclusion criteria (Hanslik 2011; Saxena 1997). As regards generalizability of the findings, all relevant types of patients and interventions were included. However, whereas arterial thrombotic events were reported in both trials, bleeding and other complications associated with the use of heparin were not reported in one of the studies.

Quality of the evidence

Overall, the quality of evidence is moderate for all outcomes due to imprecision of results and the potential limitations in the design and implementation of one of the available studies (Summary of findings for the main comparison). Indeed, one of the included trials (Saxena 1997) did not adequately describe the methods utilized for allocation concealment and blinding, thus severely limiting the possibility of assessing its selection, detection, and performance bias. As to the second study (Hanslik 2011), there were two potential problems. Firstly, the anesthetist who was taking care of the patient was aware of the allocation of the patient, which may raise concerns regarding bias. However, the anesthetist did not assess any of the outcomes thus minimizing the possibility of biasing the results. Secondly, the effect of attrition as the incidence of thrombosis was not assessed in slightly over 10% of patients in one of the arms (low-dose) and follow-up US was missing in 5.8% of patients in the other arm (high-dose). The reasons were unlikely to be related to the outcome (that is, the patient was inadvertently sent home without being assessed and the parents declined returning to the institution for US assessment, or the personnel were not available to perform the US assessment in time). Patients who did not have an US were not analyzed. Missing data were not imputed and, therefore, the study did not follow intention-to-treat (ITT) analysis as per its definition (Higgins 2008).

Importantly, the two studies differ in the detection method used to diagnose arterial thrombosis. Whereas Saxena 1997 used clinical assessment, which is currently widely used in clinical practice, Hanslik 2011 used Doppler US. Interestingly, Hanslik 2011 also recorded absent or weak pulses in their cohort. In their study, two of the six patients with arterial thrombosis as per Doppler US assessment had absent pulses, and four of the six had weak pulses; conversely, five of 196 patients without arterial thrombosis in their Doppler US had absent pulses, for a sensibility of clinical assessment in their practice of 33% (if considering only absent pulses) and specificity approaching 98%. It should be pointed out that a different study (Kulkarni 2006 ) that performed pre- and post-procedural Doppler US and clinical assessment in children of similar age (mean age 56 months) undergoing CC found clinical assessment to have 100% sensitivity and 98% specificity in the detection of arterial thrombosis.

Potential biases in the review process

Every effort was made to minimize the likelihood of missing published and unpublished trials. Our search was comprehensive and not limited to a particular language. Authors were contacted to fill in the data missing in the original publications. However, a limitation of this review is that it was not possible to collect all relevant information from the included trials, particularly safety data regarding the interventions and methodological aspects such as blinding or allocation concealment.

Agreements and disagreements with other studies or reviews

Several prospective studies have reported the incidence of arterial thrombosis in children undergoing CC (Girod 1982; Grady 1995; Ino 1988; Kulkarni 2006; Netz 1987; Ruud 2002; Vielhaber 1996; Vitiello 1998). A variety of heparin doses have been utilized, ranging from heparin flushes to 150 units/kg boluses. The incidence of arterial thrombosis has been reported to range from 3.3% to 8.6% of a total 5478 patients receiving initial doses of 150 units/kg of heparin after accessing the arterial site, followed by 75 units/kg in patients older than three months who had procedures lasting more than two hours (Ino 1988; Vitiello 1998). Two prospective studies found 0% to 0.8% frequency of arterial thrombotic events in a total of 1436 patients receiving an initial dose of 100 units/kg of heparin as a bolus, which in some cases was followed by another 50 units/kg (Girod 1982; Netz 1987). Ruud et al (Ruud 2002) found no thrombotic events in 26 CC procedures involving combined arterial and venous vascular access; their protocol consisted of 75 units/kg of heparin delivered intravenously at the beginning of each arterial catheterization; the same dose was repeated every second hour during the procedure. Kulkarni et al (Kulkarni 2006) conducted a prospective study on 120 children undergoing CC, receiving 50 units/kg of heparin. They found a 7.8% incidence of arterial thrombosis, as determined by Doppler US.

Two randomized trials compared heparin at a dose of 100 units/kg with placebo, administered either immediately after inserting the arterial catheter (Freed 1974, n = 161) or on completion of the catheterization (Rao 1981, n = 116), and reported an 8% to 11% incidence of thrombosis in the heparin arm. One randomized trial compared 100 to 150 units/kg of heparin in 60 children weighing less than 10 kg (Bulbul 2002). While no cases of pulse loss were found among 30 children receiving 100 units/kg of heparin, three out of 30 patients (10%) receiving 150 units/kg were reported to have decreased pulses post-procedure. This study was not included since current pediatric antithrombotic guidelines suggest the use of a UFH dose of 100 units/kg as a bolus, compared with a 50 units/kg dose (Monagle 2012), which results in the comparison between low (< 100 units/kg) versus high doses (≥ 100 units/kg), as defined by our review, in a more meaningful clinical comparison.

In summary, prospective studies and randomized trials that used a wide range of prophylactic doses of heparin reported a 0% to 11% incidence of arterial thrombosis in pediatric patients post-CC. Studies using initial doses of 50 to 75 units/kg of heparin reported an incidence of thrombotic events that ranged from 0% to 7.8%, suggesting that lower heparin doses might not be accompanied by an excessive number of arterial thrombotic events when compared to higher doses. These findings are in keeping with the results of the trials included in this review.

Lastly, two studies (Grady 1995; Newall 2010) compared the in vitro effect of heparin doses in children undergoing CC. In a prospective cohort of 36 pediatric patients (aged 1 month to 19.5 years) undergoing CC, Grady et al (Grady 1995) investigated the heparin effect reflected by fibrinopeptide A (FPA) as a surrogate for anticoagulation efficacy and by activating clotting time (ACT), measured in blood samples. Three patients received only heparin flushes, which failed to suppress FPA or to cause enough anticoagulation reflected by a prolonged ACT; 15 patients received a bolus of 50 units/kg of heparin and 18 patients received a bolus of 100 units/kg. Heparin doses directly correlated with suppression of FPA and ACT prolongation, with higher heparin doses having a superior in vitro anticoagulant effect. Most importantly, above a given prolonged ACT (> 200 seconds) suppression of thrombin activation, reflected by FPA suppression, was successfully achieved irrespective of the initial heparin dose. In a more contemporary prospective study, Newal et al (Newall 2010) explored the age-dependent effect of heparin measured by different laboratory assays in 56 children (aged 0 to 16 years) who underwent a cardiac angiogram. The researchers showed that the heparin concentration was age-dependent, demonstrating that at a similar dose of heparin (units/kg) the heparin concentration attained is higher in older children; and this age-dependent difference is attenuated over time after the initial heparin bolus. Moreover, the inhibitory effect of heparin on either activated factors II (FIIa) or FXa also varies with age. In summary, the first study confirmed the progressively higher capacity of increasing heparin doses to promote anticoagulation, also suggesting that if a minimal level of inhibition, reflected by the ACT, is achieved then a further increase in the heparin dose is likely to be irrelevant. The second and more complex study goes beyond these findings suggesting that heparin dosing for children undergoing CC could, in fact, be based on the patient's age rather than on one initial standardized dose of heparin. The main challenge regarding the interpretation of both studies pertains to their translational nature, in that while in vitro studies may provide a physiological plausible mechanism to explain potential clinical differences such findings ought to be confirmed by prospective clinical studies.

Authors' conclusions

Implications for practice

Although the two studies included in our review reported no evidence of a difference in the incidence of arterial thrombosis between arms, there are significant limitations such as the unclear risk of bias in one of the studies due to the lack of information provided and the limited sample size of both studies (see below). Furthermore, information about the incidence of bleeding events was incomplete in one of the trials. Interpretation of the results of these trials should combine data on both the efficacy (that is, incidence of arterial thrombosis) and safety (that is, incidence of bleeding events) of the interventions in such way that potential benefits of each intervention could be considered in the context of its side effects. There is a lack of safety information in one of the trials.

Implications for research

Despite the fact that one of the studies was designed as a superiority trial, a non-inferiority design is also a suitable option. That is, lower doses of heparin may not be significantly superior in terms of the efficacy outcome but are perhaps safer. We therefore estimated the sample size for a non-inferiority trial comparing high versus low doses of heparin, as follows. Given that the reported incidence of arterial thrombosis is approximately 6% to 8%, a trial evaluating non-inferiority of 50 units/kg (intervention) in comparison to 100 to 150 units/kg of heparin (control) would require at least 775 patients per group to achieve 80% power to detect a non-inferiority margin difference between the group proportions of 3%. The non-inferiority margin was established at 3% because this figure is to close the frequency range of thrombotic events reported in the pediatric literature, and is thus deemed an acceptable risk. The reference group proportion was assumed to be 9% under the null hypothesis of inferiority. The power was computed for the case when the actual treatment group proportion was 6%. The significance level of the test was targeted at 0.05. Therefore, a larger randomized trial will be required to answer the question of this review.

Moreover, the similarity between the results of studies included in the review and those of the non-randomized prospective studies included in the Discussion section strengthens the case for conducting a proper RCT.

Future studies should take into account the age-dependency of hemostasis and drug metabolism and researchers should ensure adequate representation of younger patients in the study. Lastly, whereas the measurement of thrombotic and bleeding complications is an essential component of any future study, determination and correlation with laboratory monitoring parameters is equally relevant.

Acknowledgements

We acknowledge Dr Derek Stephens for his guidance in the statistical aspect of this review and Ms Elizabeth Uleryk for her expertise in developing the authors' search strategy. Dr Avila was supported by a Baxter Fellowship.

Data and analyses

Download statistical data

Comparison 1. Low UFH dose compared with high UFH dose
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Arterial thrombosis2492Risk Ratio (IV, Random, 95% CI)1.06 [0.58, 1.92]
Analysis 1.1.

Comparison 1 Low UFH dose compared with high UFH dose, Outcome 1 Arterial thrombosis.

Appendices

Appendix 1. CENTRAL search strategy

#1MeSH descriptor: [Heparin] explode all trees3993
#2*heparin*7978
#3UFH or UH816
#4calciparin*30
#5danaparoid59
#6org10172 or "org 101722"5
#7Calcilean or Hepalean or Liquaemin or Eparina or heparinic39
#8*heparan*80
#9multiparin2
#10monoparin2
#11#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #108558
#12MeSH descriptor: [Heart Catheterization] explode all trees2025
#13(heart near/5 catheter*):ti,ab,kw667
#14cardiac near/5 catheter*:ti,ab,kw1431
#15coronar* near/5 catheter*:ti,ab,kw699
#16angiogra*:ti,ab,kw8607
#17arteriogra*:ti,ab,kw561
#18(arterial or artery or brachial or radial or femoral or axillary or jugular or subclavian) near/3 punctur*:ti,ab,kw207
#19(arterial or artery or brachial or radial or femoral or axillary or jugular or subclavian) near/3 catheter*:ti,ab,kw1155
#20#12 or #13 or #14 or #15 or #16 or #17 or #18 or #1912202
#21(paediatric or pediatric or child* or infant or newborn or adolescen* or youth or teen*):ti,ab,kw141547
#22MeSH descriptor: [Adolescent] explode all trees70756
#23MeSH descriptor: [Child] explode all trees122
#24MeSH descriptor: [Infant] explode all trees12348
#25#21 or #22 or #23 or #24141547
#26#11 and #20 and #25 in Trials (Word variations have been searched)50

Appendix 2. Authors' EMBASE search strategy

Ovid Embase 1980-Nov 6, 2013

  Searches Results
1heparin/110647
2(heparin or heaprinate or clarin or contusol or disebrin or eleparon or elheparin or elheparon or epiheparin or "gag 98" or gag98 or "hep lock" or hepalean or heparinate or heparine or (heparinic adj2 acid) or heparitin or hepcon or hepsal or "lipo hepin" or lipohepin or liquaemin or menaven or monoparin or mucoitin or multiparin or noparin or panheparin or panhepin or panheprin or praecivenin or pularin or thrombareduct or "thrombo vetren" or thromboliquin* or thrombophlogat or thrombophob or vetren or vister).mp.148099
31 or 2148099
4heart catheterization/45028
5((heart or cardiac) adj5 catheter*).ti,ab.27227
64 or 556574
73 and 62615
8ct.fs.476250
9clinical trial/889560
10phase 1 clinical trial/25996
11phase 2 clinical trial/42373
12phase 3 clinical trial/17475
13phase 4 clinical trial/1507
14controlled clinical trial/406276
15randomized controlled clinical trial/1
16multicenter study/114872
17meta analysis/76829
18random*.ti,ab.854611
19factorial*.ti,ab.21957
20(doubl* adj2 dummy).ti,ab.2450
21((Singl* or doubl* or trebl* or tripl*) adj5 (blind* or mask*)).ti,ab.159292
22(RCT or RCTs or multicent* or placebo* or metaanalys* or (meta adj5 analys*) or sham or effectiveness or efficacy or compar*).ti,ab.4933036
23(crossover* or "cross over*" or "cross-over*").ti,ab.68334
24(crossover* or cross over* or cross-over*).ti,ab.68334
25(Singl* adj blind*).ti,ab.14036
26(doubl* adj blind*).ti,ab.141360
27assign*.ti,ab.233714
28allocat*.ti,ab.80419
29volunteer*.ti,ab.174272
30crossover procedure/38843
31double blind procedure/118442
32single blind procedure/18456
33or/8-325910068
347 and 33983
35exp animal/ or nonhuman/ or exp animal experiment/20160156
36exp human/14963160
3735 and 3614963160
3835 not 375196996
3933 not 384670841
407 and 39965
41limit 40 to (infant <to one year> or child <unspecified age> or preschool child <1 to 6 years> or school child <7 to 12 years> or adolescent <13 to 17 years>)78

Appendix 3. Authors' MEDLINE search strategy

Ovid Medline 1946- Nov 6, 2013

  Searches Results
1heparin/49389
2(heparin or heaprinate or clarin or contusol or disebrin or eleparon or elheparin or elheparon or epiheparin or "gag 98" or gag98 or "hep lock" or hepalean or heparinate or heparine or (heparinic adj2 acid) or heparitin or hepcon or hepsal or "lipo hepin" or lipohepin or liquaemin or menaven or monoparin or mucoitin or multiparin or noparin or panheparin or panhepin or panheprin or praecivenin or pularin or thrombareduct or "thrombo vetren" or thromboliquin* or thrombophlogat or thrombophob or vetren or vister).mp.85915
31 or 285915
4heart catheterization/ or Catheter Ablation/59019
5((heart or cardiac) adj5 catheter*).ti,ab.21292
64 or 570562
73 and 6753
8randomized controlled trial.pt.388634
9clinical trial, phase i.pt.16138
10clinical trial, phase ii.pt.26806
11clinical trial, phase iii.pt.10109
12clinical trial, phase iv.pt.990
13controlled clinical trial.pt.89809
14meta analysis.pt.51298
15multicenter study.pt.181689
16randomized controlled trials as topic/102201
17controlled clinical trials as topic/5330
18clinical trials as topic/175144
19clinical trials, phase i as topic/4461
20clinical trials, phase ii as topic/6619
21clinical trials, phase iii as topic/7144
22clinical trials, phase iv as topic/231
23multicenter studies as topic/16336
24meta analysis as topic/14079
25randomized controlled trials/ or random allocation/ or double-blind method/ or single-blind method/316994
26random*.ti,ab.679086
27(doubl* adj2 dummy).ti,ab.2001
28((Singl* or doubl* or trebl* or tripl*) adj5 (blind* or mask*)).ti,ab.130714
29(RCT or RCTs).ti,ab.17561
30multicent*.ti,ab.85372
31placebo*.ti,ab.161286
32drug therapy.fs.1763008
33trial.ti,ab.356512
34groups.ab.1272819
35(sham or effectiveness or efficacy or compar*).ti,ab.3855378
36or/8-356179817
37exp animals/ not humans/4054690
3836 not 374994314
397 and 38449
40limit 39 to "all child (0 to 18 years)"57
41(pediatric* or paediatric* or child* or infan* or teen* or adolescen* or youth* or girl* or boy or boys).mp.3150078
4239 and 4157
43 40 or 42 57

Appendix 4. Virtual Health Library search strategy

(tw:(cardiac)) OR (tw:(heart)) AND (tw:(children)) OR (tw:(child)) OR > (tw:(pediatric)) OR (tw:(paediatric)) OR (tw:(pediatrics) OR > (tw:(paediatrics)) AND (tw:(heparin)) AND (tw:(randomized)) OR (tw:(randomised)) AND > (tw:(catheterization)) OR (tw:(catheterisation)) OR (tw:(catheterism))

Appendix 5. World Health Organization International Clinical Trials Registry Platform search terms

Cardiac cathererization

Appendix 6. Clinical Trials search terms

Cardiac catheterization, heparin

Appendix 7. Current Controlled Trials search terms

Cardiac catheterization, heparin

Appendix 8. Disseration databases search terms

Cardiac catheterization, heparin

Appendix 9. Grey literature search terms

Cardiac, heparin

Appendix 10. Glossary of medical terms

AnaphylaxisAcute and severe hypersensitive response to a formerly encountered substance, characterized by manifestations that can vary from respiratory and skin symptoms to shock and death
AnticoagulantSubstance, such as heparin, used to prevent the formation of a clot (prophylaxis against clot formation) or its further growing if it is already present (treatment of clots)
ArteriographyImaging technique used to visualize the structure of arteries by injecting an opaque dye into the vessel and taking an x-ray
Cardiac catheterizationMedical procedure performed by inserting a flexible tube or catheter into a blood vessel in the arm or leg. The catheter is progressed to reach the heart where it is used for diagnosis or treatment of heart conditions
ClaudicationImpairment in walking secondary to pain and weakness, in turn due to abnormal blood supply
Congenital heart diseaseMedical condition related to abnormal or incomplete development of the structure of the heart or great vessels, present at birth
Doppler ultrasoundImaging technique that uses the Doppler effect to evaluate movement of liquids (e.g., blood flow in vessels). See Ultrasound
Hemodynamic instabilityClinical state in which support (mechanical or pharmacological) is required to maintain blood flow
Hemostasis/Hemostatic reactionComplex mechanism that can be triggered by a vessel injury and, under normal circumstances, leads to arrest of the blood-flow thus stopping bleeding
HeparinType of blood-thinner. See Anticoagulant
Heparin-induced thrombocytopeniaParadoxical reaction to heparin administration characterized by decrease in the number of circulating platelets and high risk of developing clots
Muscle wastingMuscular atrophy
Systemic anticoagulationCircumstance in which anticoagulants are given in such a way that they will reach the entire body through the bloodstream
Thrombosis (arterial or venous)Refers to the formation of a blood clot inside a vessel that interferes with blood flow
UltrasoundImaging technique that uses sound waves to evaluate different body structures
VenographyImaging technique used to visualize the structure of veins by injecting an opaque dye into the vessel and taking an x-ray

Contributions of authors

Dr Avila coordinated this review and performed data collection, methodological appraisal, data management, and wrote the final manuscript. Dr Shah collaborated as the methodological expert on the design and conduct of the review, and contributed to performing the critical review of the manuscripts. Dr Brandao was responsible for the design phase of this study (research idea), collaborated with the screening and quality assessment of studies as second review author, and critically reviewed the original manuscripts.

Declarations of interest

None known

Sources of support

Internal sources

  • No sources of support supplied

External sources

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

    The Cochrane PVD Group editorial base is supported by the Chief Scientist Office.

Differences between protocol and review

None

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Hanslik 2011

MethodsRCT + cohort study
Participants

227 patients aged 0 - 20 years were enrolled, of whom 137 were randomized and 90 followed in the cohort study (age: 0 - 19 years, median 5.6 years, 23% infants)

Inclusion criteria:

Patients requiring diagnostic or interventional cardiac catheterization in whom written informed consent was obtained

Exclusion criteria:

"Exclusion criteria for the RCT were pre-existing anticoagulation or antiplatelet therapy"

Eleven patients did not have their follow-up ultrasound performed due to early discharge from hospital

InterventionsThe high-dose heparin arm received a bolus of 100 units per kg body weight heparin at the time of vascular access, followed by a continuous infusion of 20 units/kg/h standard heparin for children > 1 year, or 28 units/kg/h for infants during the procedure. The low-dose heparin arm received a bolus of 50 units per kg body weight heparin at the time of vascular access, and an additional bolus of 50 units/kg every 2 h during the procedure
Outcomes

Primary efficacy outcome was thrombosis at the puncture site diagnosed by vascular ultrasonography, performed before CC and within 48 h after CC

"The primary safety outcome was bleeding at the puncture site or other locations. Severity of bleeding was documented by the need for prolonged compression of punctured vessels or the need for medical therapy (protamin infusion, red cell transfusion)"

"Secondary study outcomes were clinical symptoms and signs of thrombotic events at the access site (palpation of femoral and distal leg pulses, skin temperature and skin color and swelling of the legs)"

Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low risk"Randomization sequences were computer generated by an independent statistician"
Allocation concealment (selection bias)Low risk"Randomization codes were allocated using sequentially numbered, opaque sealed envelopes. Envelopes were opened by an assistant in the CC laboratory, and the allocated UFH protocol presented to the anesthetist who was responsible for UFH administration via a peripheral vein"
Blinding of participants and personnel (performance bias)
All outcomes
Low risk"Patients, physicians performing the CC, those performing the clinical outcome assessment or vascular ultrasound and the study coordinators were blinded to treatment allocation. Anethesist was not blinded and was involved in the care of the patient, but not in the assessment of outcomes"
Blinding of outcome assessment (detection bias)
All outcomes
Low riskPhysicians "performing the clinical outcome assessment or vascular ultrasound were blinded to treatment allocation"
Incomplete outcome data (attrition bias)
All outcomes
Low riskFour of the 68 patients allocated to the high-dose UFH arm, and 7 of the 69 patients allocated to the low-dose UFH arm were lost to follow-up (lack of ultrasound assessment). Total attrition: 8%. As per the investigators report, the lack of outcome assessment in these 11 patients was unlikely to be related to the outcome and overall balanced across groups
Selective reporting (reporting bias)Low riskAll pre-specified outcomes were reported
Other biasLow riskThe trial was stopped early due to an incidence of the outcome which was lower than expected in an interim analysis. Interim analysis was, however, initially done without opening the randomization code

Saxena 1997

  1. a

    CC: cardiac catheterization

    RCT: randomized controlled trial

MethodsRCT
Participants

366 patients aged 17 days to 11 years old (mean age 39.5 months, 24.9% infants)

Inclusion and exclusion criteria were not provided.

Interventions50 units/kg or 100 units/kg as a single bolus, administered immediately after arterial puncture
OutcomesThe presence of thrombosis was defined by clinical findings (lack of palpable pulses at 6 hours post-cardiac catheterization)
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom numbers were used
Allocation concealment (selection bias)Unclear riskNot enough information available
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot enough information available
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot enough information available
Incomplete outcome data (attrition bias)
All outcomes
Low riskOutcome reported in same number of patients that were randomized
Selective reporting (reporting bias)Unclear riskNot enough information available
Other biasUnclear riskNot enough information available

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
  1. a

    ACT: activated clotting time
    CC: cardiac catheterization
    RCT: randomized controlled trial

Bulbul 2002RCT only comparing high doses of heparin (100 to 150 units/kg)
Butt 1987RCT comparing the effect of heparin, saline solution and different infusion rates in the patency of peripheral arterial lines in patients admitted to the pediatric intensive care unit
Chen 2013Prospective study. The authors studied laboratory endpoints in patients undergoing catheterization, comparing those receiving heparin to patients not receiving anticoagulants
Downs 1974Prospective study conducted in adult patients requiring arterial catheters for a prolonged period of time in the intensive care unit
Freed 1974RCT comparing heparin to saline solution
Girod 1982Prospective study: "Three different methods of systemic heparinization were compared to determine whether one is superior in preventing thrombosis". All three protocols included initial 100 Units per kilogram, followed or not by 1 or 2 boluses of 50 units per kilogram after
Grady 1995Prospective cohort. Not randomized (the author contacted to clarify this point)
Hollinger 2009Review of the topic. No original patient data
Kavanagh-Gray 1978Prospective study, alternate patients received heparin vs saline solution. Median age 40 years
Kim 2010Prospective study investigating laboratory endpoints (heparin monitoring)
Kulkarni 2006Prospective study analyzing the frequency of post-CC vascular complications. All patients received the same dose of heparin
Netz 1987Prospective cohort. "After the femoral artery or vein had been punctured and the catheter inserted, 100 IU heparin/kg body weight were administered."
Phillips 2010Retrospective review of post-procedural complications
Rao 1981RCT. Intra-arterial heparin or placebo were administered after CC
Roschitz 2002Prospective cohort comparing heparin to low molecular weight heparin
Roushdy 2010Prospective cohort studying factors related to vascular access complications; no doses of heparin were compared
Schroeder 2010RCT involving children after undergoing cardiac surgery. The authors compared continuous infusion of heparin at very low doses (10 units per kilogram hour) to placebo
Zeevi 1999Prospective cohort, "Heparin bolus of 30 – 100 IU/kg body weight was administered at the catheterizing physician discretion, aiming to keep the ACT level above 200 sec."

Ancillary