Evidence-based summary of preventive care for central venous access device-related thrombosis in hospitalized children

In this study, search terms of "venous thromboembolism, venous thrombosis, totally implantable venous access port, central venous catheters, peripherally inserted central catheter, central venous access device" were employed to retrieve the following guidelines and professional association websites: ① guideline websites: Guidelines International Network (GIN), Registered Nurses' Association of Ontario (RNAO), National Guideline Clearinghouse (NGC), National Institute for Health and Clinical Excellence (NICE), Joanna Briggs Institutem (JBI), Scottish Intercollegiate Guidelines Network (SIGN), New Zealand Guidelines Group (NZGG), British Medical Journals (BMJ), UpToDate, and Medlive guide; ② professional association websites: Infusion Nursing Society (INS), International Union of Angiology (IUA), International Society on Thrombosis and Haemostasis (ISTH), American Society of Hematology (ASH), American Society of Pediatric Hematology/Oncology (ASPHO) and Association of Women’s Health, Obstetric and Neonatal Nurses (AWHONN). The search of the guidelines and professional association websites was completed from July 18, 2021 to July 22, 2021.

We searched the Cochrane Library, PubMed, EMBASE and Web of Science English databases, as well as VIP, Wanfang and CNKI Chinese databases. The specific retrieval strategy was formulated according to the characteristics of each database. The time limit for database retrieval was from the establishment of each database to July 17, 2021, and the retrieval strategy was attached in Supplemental Material S1.

The inclusion criteria of the literature were as follows: (1) Published guidelines, systematic reviews and expert consensus; (2) Children (≤ 18 years old) with CVAD catheterization included in these studies; (3) Recommendations on risk factor assessment, prevention and management of CRT in children; (4) Clear recommendation; (5) Expressed in Chinese or English; (6) The revised guidelines should be incorporated into the latest version.

The exclusion criteria of the literature were as follows: (1) Guideline interpretation, translated versions of foreign guidelines and repeated guidelines; (2) Meeting summaries, draft guidelines, discussion drafts, and other relevant documents of guidelines; (3) Clinical practice implementation and application research of the guidelines; (4) Evidence list without methodological description; (5) Conference report, normative document and government draft; (6) Systematic reviews at the stage of the research plan or included in the guidelines; (7) No access to the full text.

According to the inclusion and exclusion criteria of the literature, two members of the research group who have received evidence-based nursing training independently screened the literature by reading the title, abstract and full text. If there was any disagreement, the two researchers should negotiate to resolve it. If they still cannot reach an agreement, the third researcher should be asked to step in and give his/her opinions. The data extraction table was made by the following items: articles (type), author/institution/organization, year (published/last updated), country, source (published institution/organization) and tools for evidence grading and recommendation strength.

The quality of the included articles was evaluated independently by three members of the research group trained in evidence-based nursing through the evaluation tools. If there was a conflict of opinion, the fourth member would be asked to make the judgment.

The evaluation of the Guidelines was conducted by the Appraisal of Guidelines for Research and Evaluation II (AGREE II) [16]. The AGREE II covers six aspects: scope and object, participant, rigor, clarity, application, and independence. Based on the standardized scores in each field, the overall quality of the guidelines was evaluated, and whether to recommend the guidelines was formulated by the 3-level standards. The criteria are as follows [17, 18]: if the number of fields with a standardized score < 30% was greater than or equal to three, the guideline is considered to have poor evidence quality and is not recommended temporarily (Grade C); if there were fields with standardized scores < 60%, and the number of fields with scores ≥ 30% was greater than or equal to three, the guideline needs to be modified and then it is recommended (Grade B); if standardized scores in the six fields were all ≥ 60%, the guideline does not need to be modified and can be directly recommended (Grade A).

The systematic review was assessed by a critical appraisal tool for systematic reviews (AMSTAR 2) scale [19]. In the AMSTAR 2 scale, there are 16 items (items 2, 4, 7, 9, 11, 13 and 15 are critical items and the remaining are noncritical items) and 3 evaluation indexes ("Yes", "Partly yes", and "No"). Based on the criticality and evaluation results, the systematic review study was graded with "confidence", and there were four "confidence" quality levels, namely critically low-quality, low-quality, intermediate-quality and high-quality. The detailed classification of four "confidence" quality levels is as follows [19, 20]: if there is more than one key item with or without non-key items and does not meet the standard, the systematic review is rated as critically low; if only there is one key item accompanied by or not accompanied by non-key items and does not meet the standard, this systematic review is rated as low grade; if there are more than one non-key items and does not meet the standard, this systematic review is rated as intermediate; if there is less than one or equal to one noncritical item and does not meet the standard, this systematic review is rated as high.

The Expert Evaluation Standard (2016) of the JBI Evidence-based Health Care Center in Australia was used to evaluate studies based on expert opinions and professional consensus. The evaluation standard included 6 items, and the evaluation grade of each item was "yes", "no", "unclear" and "not applicable" [21]. The inclusion criteria of the literature were as follows [22]: at least four of the six items were rated "yes". The exclusion criteria of the literature were as follows: at least two of the six items had an evaluation grade of "unclear" or included an evaluation grade of "not applicable".

Content analysis was used to analyze the included articles. Taking the included articles as analysis samples, the main evidence of CRT preventive care in children was summarized, and the topic/sub-topic of the evidence was determined. In the process of extracting recommendations, when there were conflicting evidence conclusions from different sources, this study complied with the following principles [23]: evidence-based outcome, high-quality evidence, the latest published authoritative literature and guidelines authorized by Chinese institutions were preferred. Two trained members of the research group independently evaluated the included studies. In case of conflicting opinions, the third researcher made the final decision on inclusion or exclusion.

The recommendation level of the evidence follows the original rating level in the studies. If there are repeated recommendations in multiple guidelines, the guideline with the highest score will be selected based on the quality evaluation result of AGREE II scale, and the recommendation of evidence level in the guideline shall prevail. If the level and recommendation level of evidence in the original literature is not determined, the evidence will be evaluated by the JBI grading of evidence and recommendation system (2014 edition) [24]. According to the source of evidence and the type of literature, the evidence level of the included studies was pre-graded into five levels from Level 1 to Level 5. According to JBI feasibility, appropriateness, mean-fullness, and effectiveness (FAME) structure, the recommendation strength of evidence was determined to be strong recommendation or weak recommendation.

In this study, relevant experts were invited to participate in online and offline meetings to collectively discuss the best evidence extracted. Ten experts from the Xiangya Nursing School and Xiangya Hospital of Central South University from different fields participated in this study, covering CVAD placement care, pediatric nursing, cardiovascular surgery nursing, nursing management, evidence-based nursing, nursing education, cardiovascular surgery and pharmacy. Xiangya Hospital is one of the China’s top tertiary class A general hospitals, integrating medical, educational and scientific research. JBI Xiangya Center for Evidence-Based Practice and Healthcare Innovation was established in 2019, serving as an excellent platform for evidence-based nursing research in this study. The purpose of expert consultation is to analyze the pros and cons and obstacles of nurses' application of evidence based on medical situation in China and the wishes of the children's families, improve relevant evidence or remove inapplicable evidence according to clinical applicability, and finally localize the best evidence.

The SPSS 20.0 was used for statistical analysis. The basic characteristics of the included articles were analyzed descriptively. In the quality evaluation process of the guidelines, the intraclass correlation coefficient (ICC) ranging from 0 to 1.0 was used to evaluate the consistency of evaluation results among the three members of the research group [25]. When the ICC value is less than 0.40, it indicates that there is poor consistency among study members regarding the quality evaluation results of the guideline. If the ICC value ranges from 0.40 to 0.75, the consistency among the three members is fair. If the ICC value is greater than 0.75, the consistency among the three members is high [26]. The standardized score for each area of the guideline can be calculated by the following equation: (actual score—minimum possible score)/(maximum possible score—minimum possible score) × 100%. Then, the mean, minimum and maximum values were used to describe the standardized scores in each field. The general information of experts was analyzed by mean ± standard deviation or frequency (%), and the expert consultation results were evaluated by the authority degree of expert opinions. The individual authority coefficient of the expert was calculated as: (academic level + judgment basis + familiarity) /3. The overall authority coefficient of experts was calculated by the ratio of the sum of experts' individual authority coefficients to the number of experts [27].

A total of 642 articles were initially retrieved. Based on the inclusion and exclusion criteria, 7 evidence-based guidelines, 7 systematic reviews and 9 expert consensuses were eventually included. Figure 1 shows the process of literature selection. Table 1 presents the basic features of the literature.

A total of 91 evidence were extracted and summarized in the first draft by using the method of content analysis. Evidence grading and recommendation strength were carried out in accordance with the principle of setting the evidence and recommendation levels in advance. Expert consultation was further carried out to localize the summarized evidence. The average age of experts was (41.40 ± 7.34) years old, and the average working years were (19.30 ± 12.00) years (Supplemental Table S3). The individual authority coefficient of 10 experts in this study was > 0.70, and the overall authority coefficient of the experts was 0.815, indicating that they have highly authoritative and the research results were acceptable (Supplemental Table S4). We revised and localized the items according to suggestions from experts. Finally, 68 best evidence were obtained from 14 topics, including personnel qualifications and quality management, pediatric patient selection, risk assessment, CVAD selection and use, tip position, catheter maintenance, basic prevention, drug prevention, imaging examination, health education, nursing records, follow-up, CVAD removal and others, as shown in Table 4.

In this study, the best evidence conforming to the Chinese clinical situation for CRT preventive care in hospitalized children was summarized through comprehensive retrieval, systematic evaluation and expert analysis of relevant evidence-based resources. This provides a theoretical basis for Chinese pediatric nurses to strengthen the management of CRT in hospitalized children.

Articles 1–5 of the evidence summarize how to guarantee the personnel qualifications responsible for the management of CVAD, three of which are strong recommendations. It is necessary to construct a professional team for intravenous therapy to promote the establishment and implementation of standardized evidence-based practice protocols and policies [33]. The characteristics of different hospitals should be considered, such as the differences in hospital level, type (general hospital or specialized hospital), and the abilities of members of the intravenous therapy team, and then the training program should be developed in accordance with the characteristics of the hospital and the stage of the intravenous therapy team members. It is suggested to improve the quality control responsibility by establishing an evaluation index system, supervision system and improvement system for intravenous therapy [33].

Articles 6–7 provide a summary of the topic. CVAD is appropriate for individuals who require reliable long-term vascular access and are unable to utilize the peripheral venous route because of small, immature veins, poor tolerance of repeated venipuncture in children, frequent need for access and/or a caregiver’s inability to master peripheral vein access [41]. Therefore, CVADs can only be placed in a lien if the child has a therapeutic need. CVAD requires meticulous adherence to sterile technique, and the family’s capability and commitment to meet this requirement must be carefully evaluated. It is essential that caregivers must be able to monitor for and recognize complications, such as infection, thrombosis and mechanical problems [41].

Five strong recommendations are included on this topic. The majority of pediatric patients with CRT are asymptomatic [48]. The systematic review conducted by Neshat-Vahid et al. has shown that among the included 16 studies of 1279 pediatric patients with CVCs, the incidence of symptomatic CRT is only 5% [49]. Therefore, a high index of suspicion regarding the presence of CRT should be maintained for it is usually asymptomatic [33, 41]. It is recommended to assess the risk of CRT in children [43]. Multiple risk factors have been associated with a higher risk of CRT [33, 39, 40]. Routine thrombotic screening prior to initial CVAD placement is not recommended unless the patient has ever experienced a thromboembolic event unrelated to the catheter [41]. The screening for thrombophilia depends on the following clinical circumstances [50]: ①If the child's first VTE episode is associated with CVAD, screening for thrombophilia is not recommended. ②If the child has recurrent VTE (including recurrent CRT), a thrombophilia screening is recommended. ③For individuals with a clear family history of VTE or thrombophilia, e.g., a first-degree relative under 40 years old with VTE, thrombophilia screening is recommended if there are other risk factors or underlying diseases contributing to the risk of thromboses, such as cancer, CVAD, trauma, or major surgery.

Articles 13–24 summarize the selection and use of CVAD from size, type, site, puncture, fixation, and time, 8 of which are strong recommendations. The appropriate intravenous therapeutic device should be selected according to the physicochemical properties of the drug, and the age, venous conditions, treatment duration, and disease status of the child [33]. On the premise of meeting medical needs, CVAD with the least number of lumens and the minimum diameter should be selected [3]. Placement of CVAD from the left or right side of the body did not significantly reduce the incidence of CRT [29, 30]. There are insufficient data to support a preference for the jugular or subclavian vein for CVAD. Femoral vein access is associated with a high risk of thrombosis in children and should be avoided if possible [29, 30]. The maximum duration that CVAD can be left in place has not been determined [41], but in children with ALL, there is no definitive evidence that delaying CVAD placement until the end of ALL induction therapy reduces the risk of CRT, and the delayed placement in children with poor venous access may also result in physiological and psychological consequences to some extent [29]. Ultrasound-guided CVAD placement significantly improves the rate of success, reduces the times of punctures, and decreases the incidence of complications [47]. Studies from the United States and Australia have demonstrated no difference in catheter failure and complications when CVADs are secured by different methods [51‐53].

Articles 25–29 are summarized on this topic, all of which are strong recommendations. When the catheter tip is closer to the right atrium, the blood flow at the location of the catheter increases, thereby rapidly diluting the drug and reducing the damage to the intima. At the same time, it also lowers the ratio of the drug liquid volume to the blood flow, and has less impact on local hemodynamics [3]. Studies have shown that a reduced risk of CRT is reduced when the catheter tip is located in the lower 1/3 of the superior vena cava or the junction of the right atrium and the superior vena cava [54, 55]. CVAD should not be used until the catheter tip position is confirmed or adjusted by ultrasound, X-ray or other imaging modality [3, 33, 41].

Articles 30–34 summarize the methods to reduce catheter dysfunction, 3 of which are strong recommendations. Catheter dysfunction plays an important role in unplanned catheter removal, including thrombotic dysfunction caused by thrombus, fibrin sheaths, and fibrin tails in the lumen (about 60%), and non-thrombotic dysfunction due to drug precipitation or mechanical problems [56]. A randomized controlled trial (RCT) [57] and a systematic review [58] showed no difference between saline irrigation and heparin irrigation in preventing CRT, reducing deposition to ensure long-term catheter use, and both were safe and effective. When a non-invasive needle is used to maintain the PORT patency, flushing with saline is followed by drug or blood administration and then saline is recommended [47]. Moreover, studies have shown a significant difference in the flushing effectiveness between the needle tip inclined plane parallel and opposite to the outlet channel when inserted into the puncture septum, the latter method is better [47].

Article 35 summarizes the basic prevention of CRT and is a strong recommendation. Non-drug strategies are encouraged to use to prevent CRT if possible. One study noted that children with malignant tumors and their families had poor compliance with PICC implants during daily fist exercises [59]. The following medical counseling games are proven to be effective in improving the basic prevention of CRT in children with ALL and their families. Due to the unhealed puncture wound in the early stage and the pain, the child is afraid of exercise. In the case of little blood leakage, the child is encouraged to moderately move the arm. Before the first maintenance of PICC, the child is encouraged to perform the "Little Tail Love Exercise" supervised by the guardians: Firstly, the child can watch the pre-recorded arm exercise video at the bedside. Secondly, the nurse demonstrates it face to face and teaches the child to do the loose-fist clenching, arm straightening-bending and ear-touching exercises, with 3 times a day and 20 actions each time. Finally, a "thumbs up" will be given to the child after completing tasks [60].

Articles 36–40 are summarized on this topic, 3 of which are strong recommendations. Currently, the preventive use of anticoagulants or thrombolytic drugs for the sole purpose to prevent CRT is not recommended [3, 13, 30, 34‐36, 38, 42, 47], but LWMH should be considered for children at high risk of thrombosis [13]. Low molecular weight heparin is associated with a lower risk of thrombocytopenia and osteoporosis than unfractionated heparin, and less interference with medication and diet than vitamin K antagonists [13]. Studies [36, 61] have shown that LMWH is not found to increase the bleeding risk, but further data are needed to demonstrate the value of CRT prevention in children with various specific diseases [3]. A meta-analysis [37] revealed that heparin-bonded catheters do not reduce the CRT and thrombocytopenia, indicating whether the CVAD is bonded with heparin or not is not significant. It is recommended to use t-PA to restore the patency of blocked CVAD, and an ultrasound should be performed if two doses of local thrombolysis make no difference in the patency [32].

Articles 41–42 are summarized on this topic, one of which is rated as strong recommendation. When a CRT is suspected, Ultrasound Doppler is preferred, which can indicate the location and extent of CRT and infer the thrombus freshness based on echo intensity [62]. However, the use of ultrasound to screen for CRT indiscriminately is not recommended [63]. CT or MRI may be an option if Doppler is not available or the results are unreliable or suspected to be false negatives [13]. Thrombosis in the vena cava, common iliac vein, subclavian vein, and nameless vein, as well as coexisting extravascular compression factors like tumors and thoracic outlet compression, can be clearly diagnosed by CT or MRI [64].

Articles 43–48 are summarized on this topic, and only article 43 is rated as strong recommendation. Healthcare providers should teach children and caregivers about the purpose of administration, the use of infusion devices, and the risk of potential complications, and avoid excessive use of medical terminology in health education [65]. It is important to note that preschool children should not be allowed access to CVADs and that any caregiver who will assist in the maintenance of the CVAD should attend a course of training [41]. Caregiver competency should also be assessed, such that caregivers demonstrate the care of CVAD on the model and the child at least twice [41]. In order to promote health education for children and their families, an educational checklist related to intravenous therapy in hospital or at home is available, consisting of four parts [33]: (1) General information about the child (name, age, weight); (2) The main caregivers (education level, willingness to learn, religious beliefs, language communication fluency); (3) Teaching methods (oral, written, multimedia, demonstration, others); (4) Teaching contents. The medical staff should inform the children and their caregivers that if the skin at the puncture site of CVAD is swollen, the local dressing oozes blood and liquid, and the film is wet, loose or curled, they should immediately contact the medical staff and return to the hospital for treatment if necessary.

Articles 49–50 are summarized on this topic, one of which is rated as strong recommendation. A complete record regarding CVAD nursing and problems is of great significance for tracking the whole process of catheters placement and providing a clinical basis for further management and improvement of CVAD, and facilitating the data collection for nursing teaching and research [41]. Considering that electronic health records (EHRs) are commonly applied in clinical practice nowadays, and nurses have limited ability to assess the safety of venous access and intravascular catheters, the role of EHRs in strengthening symptom monitoring, risk warning and clinical decision support deserves further exploration to promote the implementation of standardized intravascular catheter surveillance [66].

Articles 51–55 are summarized on this topic, 3 of which is rated as strong recommendation. Children with CVAD catheterization often travel back and forth between their families and hospitals. Based on this particularity, pediatric nurses need to follow up with the children and their caregivers to meet their nursing needs after discharge and ensure the continuity of nursing services. For pediatric nurses, guidelines should be developed and instituted for ongoing management, tracking and follow-up in CVAD treatment centers, and the knowledge of and adherence to guidelines should be periodically assessed in all CVAD users [41]. For children, regular follow-up is recommended for children requiring long-term CVAD placement to reconsider the need for CVAD [38], and for rapidly growing children, X-rays of the catheter tip position are recommended to monitor for abnormal positioning of the tip [41]. For caregivers, their skills in using the CVAD should be reassessed at each comprehensive follow-up visit, and if the catheter becomes problematic, more frequent monitoring and enhanced care of the CVAD should be executed [41].

Articles 56–66 summarize the indications, time, and operations, of removal, and precautions in special cases, and treatment of difficulty in removal after CRT, 5 of which are rated as strong recommendations. Current guidelines do not recommend routine catheter removal [65, 67, 68]. If the catheter is still required for treatment, it should be retained and used under anticoagulation therapy [65, 67, 69]. When there is a combination of contraindications to anticoagulation, or continued progression of symptoms despite anticoagulation therapy, catheter removal should be considered. In clinical practice, however, the dependence on catheters for treatment and the feasibility of re-establishing venous access need to be assessed [3]. For patients who are highly catheter-dependent and have difficulty establishing new venous access, the value of retaining the catheter needs to be weighed against the other potential risks associated with thrombosis, and the catheter can be retained under close observation and follow-up [3]. In terms of timing of catheter removal, for catheter-associated DVT, it is recommended to be removed after a period of anticoagulation to facilitate thrombus stabilization [70]. However, for other types of CRT, the removal timing is not restricted considering the small volume of the thrombus [3].

Articles 67–68 are summarized on this topic and article 68 is rated as strong recommendation. It should be noted that CRT prevention in children cannot be separated from overall VTE prevention, especially the DVT of the lower extremities caused by CRT is greatly harmful to children. Therefore, it is still necessary to take appropriate preventive measures against the risk factors of VTE for children at a high risk of CRT [3].

In this study, the best evidence on the preventive care of CRT in hospitalized children was summarized and localized to provide pediatric nurses with an evidence-based reference in clinical practice. This study provides a theoretical basis and technical standards for pediatric nurses to strengthen the risk prevention and management of CRT in hospitalized children, and guarantees the safety of children with tubes.

Nevertheless, this study only summarizes the evidence and has not yet completed the transition from scientific evidence to clinical practice, and the evidence-based decision-making and practice are required to validate the effectiveness of the best evidence of CRT preventive care for children.