4.1. Participants, concept, and content

This study included healthcare professionals (HCP), such as nurses, medical doctors, paramedical HCPs, and vascular access specialists, who perform PIVC insertion, maintenance, and flushing processes, and laboratory professionals who conduct research in this field. The study population includes pediatric and adult patients, as well as animal models where applicable.

All types of peripherally inserted catheters, such as integrated, non-integrated, short, long, extended dwell, and midline catheters used for intravenous therapy (such as chemotherapy, medication administration, and parenteral nutrition), were included in the review. Midline catheters were not considered PIVCs but were included separately in the analysis if applicable. Studies focusing on flushing technique, flushing speed, flushing volume, flushing frequency, and flushing interval were included. Studies were excluded if they focused on the PIVC flushing or locking regimes or solutions (e.g., comparing the outcome difference between heparin and saline).

The context included clinical, laboratory, and simulated settings, including any environment used for training or research purposes, such as manikins, task trainers, and simulation technologies like virtual reality and augmented reality, as long as they relate to catheter flushing practices. In the clinical setting, both inpatient and outpatient settings were included.

4.2. Types of sources

The review included all study methodologies such as qualitative, quantitative, and mixed methods studies. The review included a variety of study designs, including non-experimental, experimental design, retrospective cohort studies, case-control studies, and simulation studies. We included evidence synthesis designs such as systematic reviews with and without meta-analysis and meta-synthesis and scoping reviews. Grey literature, such as conference papers and PhD thesis, were included.

4.3. Information sources

The search strategy was carried out following JBI’s three-phase search strategy [24]. In the first phase, an initial limited search of the flushing of PIVCs was undertaken in PubMed and Embase to identify relevant articles on the topic of interest. The free text keywords and index terms noted from these relevant articles helped further devise our search terms and develop our search strategy. PJ, JD, OH, and CD contributed to developing search terms.

The search was conducted across several databases, including PubMed, Embase, Web of Science, Scopus, and Cumulative Index to Nursing and Allied Health Literature (CINAHL). Clinical trial registries such as the Australian and New Zealand Clinical Trial Registry (ANZCTR), European Union Clinical Trials Register (EUCTR), and ClinicalTrials.gov were also searched. Grey literature was identified by reviewing the first 10 pages of Google and Google Scholar using specific keyword combinations. The search strategy included free-text keywords and MeSH terms related to PIVC and flushing techniques (see Appendix I for detailed search terms and strategy). The search process was finished in October 2024.

4.4. Selection of sources of evidence

All identified literature was imported into Rayyan, and duplicates were removed. Two research team members (JD and OH) were involved in manually screening the literature retrieved from our dedicated search strategy. OH screened 10% of the literature to assess for agreement, and JD continued screening the remaining 90%. The third member (PC) independently used the ASreview tool for study screening, stopping after 100 consecutive non-relevant papers. To ensure a comprehensive review and minimize bias, the manual and ASreview screening results were compared. The research team members discussed discrepancies between the two methods to reach a consensus.

Full texts of potentially relevant studies were reviewed independently by one reviewer (JD). The second and third reviewers (OH and PC) also reviewed 10% of the studies to resolve discrepancies and ensure consistency.

4.5. Data charting process

A standardized data extraction form (Appendix II) was used to extract data, including author(s), year of publication, country of origin, study design, study settings, study subjects, flushing techniques, flushing speed, flushing volume, flushing speed, and flushing frequency, flushing interval, and reported outcomes. One reviewer (JD) conducted the extraction work. We adopted a double-checking system with 10% of the studies, where second and third reviewers (OH and PC) verified the precision and comprehensiveness of the extracted data.

4.6. Analysis, presentation and discussion

Findings are presented using frequency counts and a narrative summary. Where appropriate, data are displayed using figures, tables, and graphs. Our discussion is divided into two sections: the first addresses the four scoping review questions, and the second highlights additional themes and findings identified during the review. The discussion further elaborates on both sections, guided by the PAGER framework [23].

4.7. Protocol amendments

Two amendments were made to our published scoping review protocol. Firstly, we limited the grey literature source to only include the first 10 pages of Google and Google Scholar with specific keyword combinations due to the large body of sources with low quality and irrelevant to the research questions. Secondly, the term PVC (peripheral venous catheter) in the scoping review protocol was replaced with PIVC (peripheral intravenous catheter) to align with standard terminology and better reflect the scope of our research.

5.1. Studies inclusion

Our search strategy retrieved 4539 sources and 1418 duplicates were removed, leaving 3121 for title and abstract screening. After two authors’ title and abstract screening, 71 studies were left for full-text screening. We subsequently conducted full-text screening, and then 39 studies met the final extracting requirements, as seen in PRISMA flow diagram (Fig 1) [25]. See Table 1 for a summary of information from the included studies.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. For a summary of information from the included 39 studies.

https://doi.org/10.1371/journal.pone.0330125.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA 2020 flow diagram.

https://doi.org/10.1371/journal.pone.0330125.g001

5.2. Research design

We categorized the research designs into several types: 4 review studies (integrative and narrative reviews) [19,57–59], 6 intervention studies (including 5 randomized clinical trials) [26,27,29–31,60], 4 cross-sectional studies (descriptive and survey-based) [32–35], 11 observational studies (retrospective and prospective cohorts) [6,36–45], 9 experimental studies (including quasi-experimental and simulation studies) [15,17,46–52], 4 mixed-methods studies [53–56], and 1 guideline [14].

5.3. Research subject and setting

The research subjects are categorized as follows: 11 studies focused on healthcare professionals, specifically nurses and midwives, as the professional discipline performing the flushing practice [32–34,40–43,45,53–55]. 15 studies focused on hospitalized patients, including infants, newborns, and adults, assessing complications or outcomes associated with flushing [5,24,25,27–29,34–37,42,54–56]. 2 studies investigated hospital-wide flushing practices at an institutional level [35,59]. 10 studies were conducted in simulated or laboratory environments to evaluate flushing techniques under controlled conditions [15,17,43,46–49,51].

5.4. Flushing techniques

10 studies did not specify the flushing techniques, only vaguely described as manual flushing or flushing catheters [29,30,33,34,36,42,43,49,54,55]. 12 studies compared different flushing techniques, such as pulsatile flushing versus continuous flushing, or steady flow flushing versus pulsed flushing flow [6,15,26,27,37–39,44,47,48,50,51]. One RCT study compared the effects of pulsatile flushing and continuous flushing. The result showed no statistical difference between pulsatile flushing and continuous flushing techniques regarding the time and type of PVC patency [27]. Other studies included intermittent flushing, continuous infusion, positive flush, and continuous and intermittent flushing combinations.

5.5. Flushing speed

The majority of the studies did not specify the flushing speed [6,15,26,29–37,39–43,45,49,54–56,60]. One simulation study mentioned a steady flow rate of 4 cm³/s with a syringe pump [15]. Another study described two different flushing methods: one group received 5 successive boluses of 1 mL each, completed in 0.5 seconds, while the other group received a single 5 mL bolus at a rate of 10 mL/min, with a flushing time of 30 seconds [48]. Other studies mentioned varying flow rates, such as 2.8, 6.8, 12.0, and 18.5 mL/s, or indicated that the flushing speed was “faster than recommended” [17,27,47,50]. Some studies focused on continuous infusion, reporting varying infusion rates such as 0.2 mL/h, 3 mL/h, 10 mL/h, 20 mL/h, and 40 mL/h [38,44,46]. With only one interventional study on flushing speed [48], there is an opportunity to develop an RCT comparing clinical outcomes of different flushing speeds.

5.6. Flushing volume

The flushing volumes are varied. 9 studies did not specify the amount [6,15,26,28,35,36,42,43,46]. The most mentioned volumes are 10 mL and 5 mL, often used alone or in comparison with other amounts like 2 mL or 3 mL. Some studies provided a range, such as 2–10 mL [6,26,28–35,38,39,44,46,47,54,55,59,60]. Two RCTs investigated the effect of flushing volume in combination with frequency [29,30]. Both compared low-volume (3 mL) versus high-volume (10 mL) flushing. The results showed no significant difference in PVC failure rates between the low- and high-volume groups. Overall, while 10 mL and 5 mL are common, there is variability and some lack specification.

5.7. Flushing frequency

We identified 13 studies that report scheduling of flushing is varied with no definite flushing frequency [5,13,24,26,32,33,38,40,41,43,44,47,53]. Further, 20 studies noted variability based on clinical circumstances and requirements, such as more frequent flushing during night shifts, after PVC insertion, before, between, and after drug delivery, or as needed [25,27–31,34,37,39,42,45,46,48–52,54]. Two RCTs also assessed flushing frequency, comparing low-frequency flushing (every 24 hours) with high-frequency flushing (every 6 hours). The findings indicated no significant reduction in PVC failure rates with more frequent flushing [29,30]. Common specific flushing frequencies were included in n = studies q24h(every 24 hours), q6h(every 6 hours), q8h(every 8 hours), and q12h(every 12 hours), often with medication administration [29–31,36,39,44,47,48,50–52,60]. Overall, the included studies have a mix of specific and variable frequencies.

5.8. Syringe used for flushing

The results regarding flushing syringe sizes, types, and prepared methods varied. A substantial portion (n = 20) of the included studies did not specify the syringe type, sizes, and prepared methods. Specific syringe sizes ranged from 2 ml to 10 ml and included 2 ml, 3 ml,5 ml, and 10 ml,. Certain studies highlighted brand-specific syringes, such as the BD Posi-Flush and BD PosiFlush XS syringes [27,43], and one study reported using a syringe pump [48].

5.9. Study outcome assessment for included studies

The included studies evaluated how various factors related to PIVC flushing influence catheter-related complications, describes as infection, infiltration, extravasation, occlusion, phlebitis, and thrombosis, as well as catheter failure. Several studies investigated the practitioner’s performing the catheter flushing in adherence to established guidelines (flushing frequency, volume, speed) [33,35,40]. A subset of studies also examined mechanical behaviors, flow characteristics of flushing techniques, and the impact of residual liquids on bacterial growth [47,50,61]. These outcome assessments collectively highlight the importance of safety, effectiveness, and protocol adherence in PIVC management.

6.1. Variation in PIVC flushing practice

The lack of a standardized protocol for PIVC flushing practices has led to significant variation across different regions and healthcare settings. Several factors, including differences in educational training, availability of medical resources, and staffing levels, influence this variation in practice [62]. Additionally, healthcare professionals may have a limited understanding of the importance of proper PIVC management, which can contribute to inconsistent practices [63,64]. Implementation studies are limited, with few studies exploring how research on PIVC flushing can be effectively translated into clinical workflows. One study evaluated errors in nursing teams’ flushing practices and developed a flushing prototype—a structured care guide to promote good flushing practices in intensive care units [53]. A single-center, stepped-wedge, cluster-randomized trial demonstrated the impact of a multifaceted intervention focused on PIVC maintenance [28]. The intervention, which included education on practice guidelines and the use of manufacturer-prepared pre-filled flush syringes, resulted in a significant reduction in PIVC failure rates—30% in the control group vs. 22% in the intervention group (risk difference −8%, 95% CI −14 to −1, p = 0.032). Importantly, the intervention also reduced total costs without any serious adverse events. These implementation studies emphasize the need to bridge the gap between research and practice, offering an opportunity for translational and implementation science to develop practical strategies that enhance guideline adherence.

6.2. The diversity in PIVC flushing techniques, speed, volume, frequency, and interval

One of the purposes for carrying out a scoping review is to identify the number of randomized controlled trials to justify the need for a full systematic review and meta-analysis [20]. In this scoping review, we included various clinical study designs, including RCT (n = 5), cohort studies (n = 4), and case-control study (n = 1). Three of the five RCT studies aimed to compare the high PIVC flushing volumes(3 ml or 10 ml) and frequencies(every 6 hours or every 24 hours) to evaluate the impact on catheter failure and complication rates [29–31]. However, significant variability existed across these studies regarding flushing techniques, volumes, frequencies, patient populations, and other factors such as PIVC gauge and insertion site. This heterogeneity causes considerable confounding, limiting the ability to draw definitive conclusions about the optimal flushing strategies. Additionally, methodological differences—such as variations in patient populations (e.g., adult vs pediatric), treatment types (e.g., antibiotics or parenteral nutrition) definitions of outcomes (e.g., catheter failure rates vs dwell time). Given these limitations, conducting a systematic review with meta-analysis would be beneficial to quantitatively synthesize the available evidence and to clarify the relationship between specific flushing parameters and catheter outcomes. Additional rigorous clinical trials are essential to determine the optimal flushing techniques, volumes, and frequencies, particularly across different patient populations and clinical treatment types.

6.3. Physiological impact and safety of different flushing techniques

The included studies include several PIVC flushing techniques, including continuous flushing, intermittent flushing, and pulsatile flushing, representing current clinical flushing techniques. Continuous flushing provides a steady flow of solution with no interval, while intermittent flushing is used at set intervals to maintain catheter patency [1]. Pulsatile flushing, which involves short bursts of solution with pauses, is recommended to enhance debris clearance from the catheter [1,65]. However, current practice guidelines advocate using pulsatile flushing primarily based on experts’ opinions with limited evidence [12]. Ferroni et al. demonstrated that pulsatile flushing significantly reduced bacterial colonization, particularly for Staphylococcus aureus, compared to continuous flushing [50]. Boord et al. found that pulsatile flushing is better for maintaining catheter patency and clearing solid deposits from the catheter walls [12]. These findings support the theoretical benefits of pulsatile flushing; however, this evidence comes from in vitro studies, limiting the generalizability of the results to clinical practice. However, an included RCT study compared the effects of continuous flushing on PIVC and found no difference between continuous flushing and pulsatile flushing regarding the time and type of PIVC patency [27]. Another prospective study compared the duration of PIVC patency between continuous flushing and intermittent flushing and found that intermittent flushing resulted in a significantly longer cannula patency duration than continuous infusion (geometric mean 47.1 vs. 35.4 h, P = 0.041). The incidence of extravasation was higher with continuous infusion (68.9% vs. 43.2%; P = 0.001), while occlusion was more common with intermittent flushing (28.4% vs. 6.6%; P = 0.002) [44]. The pulsatile flushing shows theoretical benefits in laboratory settings, its clinical advantages remain uncertain. These mixed results suggest that the optimal flushing technique may vary depending on the clinical context and patient population. Further research is needed to establish standardized guidelines for these variations and ensure the best clinical outcomes.

6.4. Lack of research on the mechanisms of PIVC flushing

The understanding of the mechanisms behind PIVC flushing is limited, and few studies have explored fluid dynamics in the PIVC flushing process. Fluid dynamics in the context of PVC flushing refers to the study of fluid flow behavior and forces within the catheter during the flushing process. This includes how the flushing saline interacts with the catheter walls, generates turbulence and creates shear forces to remove the solid deposits in the internal catheter [17]. The studies by Vigier et al. and Ferroni et al. examined pulsatile flushing but used different methods [15,50]. Vigier et al., used a transparent rectangular duct (10 cm × 1.5 cm × 0.2 cm) connected to a pump to create controlled steady or unsteady flow. A thin layer of solid deposit was formed at the bottom by evaporating a mineral powder suspension in water, simulating catheter deposits. It took 70 seconds to initiate deposit removal under steady flow conditions (4 mL/s), whereas applying an intermittent flow, with the velocity doubling at 2-second intervals, reduced this time to 25 seconds. Ferroni et al., used polyurethane catheters contaminated with a controlled concentration of Staphylococcus aureus. Two flushing techniques controlled by a pump were compared: pulsative flushing (ten 1 mL boluses of saline) and continuous flushing (a single 10 mL bolus). Catheters flushed with the pulsative method had a significantly lower median bacterial count of 524 CFU/mL compared to 1,616 CFU/mL observed in catheters flushed with the continuous method, demonstrating the higher effectiveness of the pulsative technique in reducing bacterial contamination. However, the study by Zhu et al., investigated the mechanisms of pulsatile flushing using various bolus volumes and found that higher bolus volumes (1.5 mL and 2.0 mL) within 0.5 seconds, generating peak shear rates up to 10,000 s ⁻ ¹ can significantly increase the risk of mechanical endothelial injury [17]. The range for shear rates in vivo for the vein is 20–200 s⁻¹ [65]. However, shear rates over 10,000 s⁻¹ near the vessel wall induce platelet adhesion to thrombogenic surfaces [66,67]. Shear rates over 10,000 s ⁻ ¹ near the vessel wall induce platelet adhesion to thrombogenic surfaces. This process is mediated by the von Willebrand factor (vWF), which undergoes conformational changes under high shear, exposing binding sites that facilitate platelet aggregation. Elevated shear rates can also promote platelet activation and secretion of procoagulant substances, enhancing thrombus formation [68,69]. The experimental setup allowed for precise control over variables such as flow rate, bolus volume, and interval timing, allowing for accurate results regarding the effectiveness of flushing [40]. Educational materials/ clinical guidelines typically describe pulsatile flushing as a one-second flush followed by a one-second pause with no detailed description of flow rate, bolus volume, and interval timing [1]. Furthermore, there is a lack of detailed clinical data on actual flushing practices, making it difficult to determine whether the potential benefits of pulsatile flushing are superior to continuous infusion, such as keeping vein open approaches or manual flushing at a prescribed approach, outweigh the risks of endothelial injury. No meta-analysis has been conducted to evaluate the efficiency or efficacy of different flushing methods in clinical practice. Conducting a meta-analysis of existing studies is justified, as is the design of clinical simulation studies, conducting lab-based experiments measuring endothelia injury with flushing. Such approaches are essential to establishing robust evidence for PIVC flushing practices.