Impact of different tip locations of the midline catheter on complications: a systematic review and Bayesian network meta-analysis

Introduction

Background

The midline catheter (MC) serves as an essential tool for short- and mid-term intravenous access, offering considerable clinical benefit in individuals requiring prolonged intravenous therapies (such as parenteral nutrition, pain management, or antibiotic administration) (1). As a special form of peripheral venous access (2), suitable for infusing isotonic or near-isotonic solutions. Hypertonic solutions may increase the risk of mechanical complications (such as phlebitis) (3). In recent years, the application of MCs in intravenous therapy has increased significantly (4), however, the incidence of mechanical complications (such as deep vein thrombosis and catheter occlusion) remains relatively high (5). Epidemiological data indicate that the incidence of catheter-related thrombosis (CRT) varies between 5% and 25% of cases, while superficial venous thrombosis occurs in 15% to 40% of cases (6). Thrombosis not only prolongs hospitalization and increases medical costs but may also lead to severe sequelae including chronic venous insufficiency or rarely pulmonary embolism (7). In addition, infections, catheter tip malposition, catheter leakage, and catheter dislodgement can considerably compromise treatment continuity and patient quality of life (8). MCs are primarily used for drug therapies compatible with peripheral veins (PVs) (such as antibiotics and non-irritating medications), and their safety has been validated in multiple studies. A retrospective cohort study on outpatient parenteral antimicrobial therapy (OPAT) showed that compared with peripherally inserted central catheters (PICCs), MCs had a lower risk of major complications (0.8% vs. 3.4%, P<0.001) and a similar device failure rate (9.6% vs. 12.1%). Particularly when the treatment duration is ≤14 days, MCs significantly reduce the risk of major complications [adjusted hazard ratio (aHR) =0.29] (9).

Rationale and knowledge gap

Emerging evidence suggests that the location of the catheter tip may be a key factor influencing the risk of complications. For instance, tip placement in high-flow veins such as the proximal axillary vein (AV) or subclavian vein (SV) may reduce fibrin sheath formation due to turbulent flushing, thereby lowering the risk of thrombosis. In contrast, tip placement in low-flow veins including the distal AV or the arm vein may lead to blood stasis, promoting thrombogenesis (10). However, a study on this subject reported conflicting conclusions; it suggested that tip placement in the SV or brachiocephalic vein (BV) is related to a reduced incidence of CRT (11), while other work indicates that placement in the proximal AV may elevate thrombosis risk due to localized hemodynamic alterations (12). Moreover, findings related to catheter patency are conflicting. For instance, it has been reported that tip placement in larger veins, including the BV or SV, may enhance the stability of blood flow and decrease occlusion risk (13). However, another study did not observe such significant effects (14). The inconsistency in these results may be attributed to methodological heterogeneity and limitations, including small sample sizes, varying study designs, and insufficient follow-up durations.

Objective

Although previous studies have investigated the impact of MC tip positions on catheter patency and complications (11-14), key research remains limited. Additionally, large-scale, multicenter randomized controlled trials (RCTs) that systematically assess the relative safety and effect of different tip locations are scarce. Most available evidence has been derived from single-center, retrospective studies, limiting the overall quality of the findings. Furthermore, a study (15) on this subject has predominantly depended on direct comparisons, has not incorporated indirect evidence across multiple anatomical locations, and thus can be considered incomplete assessments. Hence, this study aimed to integrate current evidence, compare the impact of various tip locations on catheter patency and the incidence of complications, and identify potential risk factors. Our findings may address the existing evidence gaps and provide more robust, evidence-based guidance for clinical decision-making. We present this article in accordance with the PRISMA NMA reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-333/rc).

Methods

This study has been registered on the PROSPERO (No. CRD420251021380).

Search strategy

Major international and Chinese databases, including PubMed, Embase, Web of Science, Cochrane Library, China National Knowledge Infrastructure (CNKI), Wanfang Data, Cumulative Index to Nursing and Allied Health Literature (CINAHL), and Chinese Scientific Journals Database (VIP), were searched from inception to March 5, 2025. Search terms included “midline catheter”, “tip position”, and “complications”. The detailed search strategy is provided in Table S1. Two reviewers (X.Y. and W.W.) independently screened titles, abstracts, and full texts, with disagreements resolved by discussion. References of included articles were manually screened to identify additional eligible studies.

Inclusion and exclusion criteria

Two reviewers (X.Y. and W.W.) independently screened all retrieved records for eligibility, resolving any disagreements through discussion. The inclusion criteria were as follows: (I) studies involving adult patients (aged >18 years) undergoing MC placement; (II) comparisons of different catheter tip positions; (III) studies reporting outcomes related to phlebitis, catheter-related embolism, CRT, or overall incidence of adverse events; and (IV) study designs including RCTs, cohort studies. Although our protocol initially also planned to include cross-sectional studies, they were excluded during the review process because such designs are methodologically inappropriate for network meta-analysis. Meanwhile, the exclusion criteria were as follows: (I) animal studies; (II) reviews; (III) case reports, guidelines, or conference abstracts; (IV) duplicate publications; (V) studies without accessible full texts; and (VI) studies not reporting relevant clinical outcomes.

Data extraction

Two reviewers (X.Y. and W.W.) independently extracted data using a predesigned form, resolving discrepancies through discussion. Extracted variables included: (I) study characteristics (publication year, first author, study design, and country); (II) patient demographics (type, sample size, age, sex ratio); (III) MC placement methods (e.g., anatomical landmark positioning, ultrasound guidance, radiographic or electrocardiographic confirmation, electromagnetic navigation); and (IV) clinical outcomes (phlebitis, CRT, catheter occlusion, and overall complications), which were defined according to standardized diagnostic and grading criteria. If data were missing, we attempted to contact the corresponding authors to request the necessary information. Detailed definitions of outcomes, intravenous therapy duration, and anatomical site classifications are provided in Appendix 1.

Outcome definition

In this study, clinical outcomes included catheter occlusion, CRT, and catheter-associated phlebitis. Each outcome was defined, diagnosed, and graded according to standardized criteria based on published guidelines and prior literature. Briefly, catheter occlusion was classified as thrombotic or non-thrombotic; CRT encompassed intraluminal thrombus, fibrin sheath formation, and deep vein thrombosis; and phlebitis was categorized as mechanical, chemical, or infectious. Detailed definitions, diagnostic criteria, and grading standards for each outcome are provided in Appendix 1.

Quality assessment

For RCTs, cohort studies, the National Institutes of Health (NIH) Quality Assessment Tool was employed to assess potential biases, using the version appropriate for each study design (16). For RCTs, the NIH Quality Assessment Tool for Controlled Intervention Studies was applied, evaluating aspects such as randomization, allocation concealment, blinding, incomplete outcome data, and selective reporting. For cohort studies, the NIH Cohort Study Quality Assessment Tool was used, covering selection bias, confounding control, measurement of exposures and outcomes, and loss to follow-up. The quality of studies was rated as good, fair, or poor according to the specific scoring criteria of each tool.

In addition to these individual assessments, the overall quality of evidence for the primary outcomes of this review was evaluated using the Confidence in Network Meta-Analysis (CINeMA) tool (17). CINeMA evaluates the quality of evidence in network meta-analysis based on five key domains: risk of bias, indirectness, inconsistency, imprecision, and publication bias. The tool allows for the classification of the evidence into four levels of confidence: high, moderate, low, and very low. These levels are determined by assessing the strengths and limitations of the studies included in the network meta-analysis. High confidence indicates strong evidence with minimal bias or uncertainty, while low or very low confidence reflects significant limitations in the evidence that may affect the reliability of the results.

Statistical analysis

All statistical analyses were conducted using R software (version 4.3.3; R Foundation for Statistical Computing, Vienna, Austria) and STATA software (version 15; StataCorp LLC, College Station, TX, USA). A Bayesian network meta-analysis (BNMA) was employed to compare the effects of different MC tip locations on complication risks. Between-study heterogeneity was assessed using the I² statistic, with I²>50% indicating substantial heterogeneity. For datasets with I²≤50%, both fixed-effect and random-effects models were fitted. Model selection was based on the Deviance Information Criterion (DIC): the fixed-effect model was preferred when the DIC difference was <5, as it offers more precise estimates with fewer assumptions; otherwise (DIC difference ≥5 or initial I²>50%), the random-effects model was selected to incorporate between-study variation.

Model parameters were estimated using Markov Chain Monte Carlo (MCMC) methods within the Bayesian framework. To mitigate dependence on initial values, 10,000 burn-in iterations were performed, followed by 40,000 sampling iterations. Four independent Markov chains were run, and convergence was evaluated using the Gelman-Rubin statistic (R̂≤1.05 indicating good convergence) alongside visual inspection of trace plots. Non-informative prior distributions were specified for all model parameters to ensure posterior inferences were predominantly driven by the observed data; specifically, the log relative risk (RR) for outcomes was assigned a normal prior distribution with a mean of 0 and a large variance of 1,000.

To further evaluate the robustness of the primary findings, sensitivity analyses were conducted using multiple approaches. First, we examined the potential influence of model choice by re-estimating the effects with a random-effects model when the primary analysis employed a fixed-effect model, and vice versa, and systematically comparing the results. Second, we assessed the impact of study design (a potential effect modifier) by repeating the analyses separately for cohort studies only and for combined datasets including both cohort studies and RCTs. Inconsistency between direct and indirect evidence within closed loops in the network was assessed using the node-splitting approach. The relative ranking of tip locations for each complication outcome was determined by calculating the surface under the cumulative ranking curve (SUCRA) values, ranging from 0% to 100%, with higher values indicating a more favorable rank (i.e., lower risk of the specific complication).

Potential publication bias was investigated using comparison-adjusted funnel plots, generated with the netfunnel command in STATA and augmented with a regression line. Symmetrical distribution of points around the regression line suggested a low risk of publication bias, while asymmetry indicated potential bias. Additionally, for pairwise comparisons involving ≥10 studies, Egger’s linear regression test was performed, with a P value <0.05 signifying statistically significant publication bias. Recognizing pathophysiological distinctions in phlebitis manifestations, a prespecified subgroup analysis differentiated mechanical and chemical phlebitis, thereby accounting for potential effect modification by etiology.

Results

Search results

Initially, 1,882 studies were obtained from eight databases and screened via EndNote (Clarivate, Philadelphia, PA, USA). After 537 duplicates were removed, 1,289 studies were excluded based on titles and abstracts. After full-text screening, five reviews, 10 conference abstracts, 19 studies without comparisons of different tip positions, nine studies unrelated to MCs, and three case reports, and one study with overlapping population data were removed. Ultimately, 9 studies were included. In addition, citation screening identified eight duplicate records already present in the screening database, which were excluded. The literature screening process is illustrated in Figure 1.

Figure 1 Literature screening process. CINAHL, Cumulative Index to Nursing and Allied Health Literature; CNKI, China National Knowledge Infrastructure; VIP, Chinese Scientific Journals Database.

Characteristics of included studies

Nine studies were included, comprising two RCTs, seven cohort studies. Among them, seven studies were from China, one from the United States, and one from Italy. The reported tip positions of MCs included the SV, distal AV, proximal AV, axillo-subclavian junction (ASVJ), brachial vein (BrV), and BV. The catheter dwell time in most studies exceeded 7 days, and participants were predominantly older adults (age >60 years). The primary study outcomes were overall complications and CRT, with catheter occlusion and phlebitis considered secondary outcomes. The characteristics of the included studies are detailed in Tables 1,2 (11-14,18-22). The original dataset used for the statistical analyses is provided in Table S2 to enhance transparency and reproducibility.

Quality assessment

Quality assessment demonstrated that two studies were rated as good and seven as fair, with no study rated as poor. The complete quality evaluation results are shown in Tables S3,S4. According to the CINeMA framework, the CINeMA assessment showed that for catheter occlusion, most comparisons performed well across all domains, with overall confidence ratings mainly at a moderate level and a few downgraded due to small sample sizes and imprecision. For CRT, most comparisons were rated as moderate confidence, with some downgraded for within-study bias and imprecision, particularly in comparisons with limited sample sizes. For phlebitis, the majority of comparisons achieved high or moderate confidence, with only a few downgraded due to imprecision. For overall complications, most comparisons performed well in all domains, with confidence levels predominantly high or low. Overall, the evidence quality across the four outcomes is relatively robust, although comparisons based on small sample sizes frequently exhibited imprecision and should be interpreted with caution. Detailed CINeMA assessments are provided in Figures S1-S4.

Assessment of model assumptions and diagnostics

Considering the comparability in study design, outcome measurement, population characteristics, and inclusion and exclusion criteria, we deemed that conducting a BNMA to synthesize quantitative evidence was appropriate. Homogeneity was assessed using the I² statistic, and overall heterogeneity across the included comparisons was low. Consistency between direct and indirect evidence was evaluated using the node-splitting approach, and no statistically significant inconsistency was detected across any closed loops. Model fit was compared using the DIC, with differences between fixed-effect and random-effects models all less than 5. Convergence diagnostics based on the Gelman-Rubin statistic (R̂≤1.05) and visual inspection of trace plots indicated good convergence for all model parameters. Details are shown in Table S5.

Incidence of overall complications

Seven studies reported the incidence of overall complications, with the tip positions covering the SV, distal AV, proximal AV, and BV, involving a total of 960 participants and 181 adverse events. The network plot is shown in Figure 2A. Given the presence of a closed loop, an inconsistency test was conducted, and the results revealed no significant inconsistency (Figure S5). A forest plot (Figure 2B) demonstrated that the proximal AV, SV, and BV were related to the notably reduced incidence of overall complications compared with the distal AV. The SUCRA rankings demonstrated that the BV was correlated with the lowest incidence of overall complications [RR =0.15; 95% credible interval (CrI): 0.01–0.74; SUCRA =85.9%], followed by the SV (RR =0.27; 95% CrI: 0.19–0.38; SUCRA =75.9%). However, the wide CrI for the BV suggests potential uncertainty, possibly due to the small sample size. The results of the pairwise comparison are illustrated in Table S6.

Figure 2 BNMA of the complication rates. (A) Network diagram. (B) Forest plot of relative effects. BNMA, Bayesian network meta-analysis; CrI, credible interval; SUCRA, surface under the cumulative ranking curve.

CRT

Nine studies reported data on CRT, with tip positions including the SV, distal AV, proximal AV, BV, ASVJ, and BrV, involving a total of 2,000 participants and 78 adverse events. The network plot is presented in Figure 3A. Given the presence of a closed loop, an inconsistency test was carried out (Figure S6), and the results indicated no significant inconsistency. The forest plot (Figure 3B) demonstrated that compared to the distal AV, the SV was associated with a statistically significant reduction in the risk of CRT (RR =0.36; 95% CrI: 0.13–0.92; SUCRA =91.0%). In contrast, the junction ASVJ was associated with a statistically significant increase risk of CRT (RR =26.88; 95% CrI: 4.74–237.30; SUCRA =0.3%). The wide CrI suggests potential uncertainty, likely due to the small sample size. As a transitional zone for vascular alignment and hemodynamics, ASVJ is prone to the formation of turbulent flow or low shear stress, which may increase the risk of thrombosis. Pairwise comparisons are illustrated in Table S6.

Figure 3 BNMA of CRT. (A) Network diagram. (B) Forest plot of relative effects. BNMA, Bayesian network meta-analysis; CrI, credible interval; CRT, catheter-related thrombosis; SUCRA, surface under the cumulative ranking curve.

Catheter occlusion

Eight studies reported outcomes related to catheter occlusion, involving a total of 960 patients and 49 catheter occlusion events. The network plot is illustrated in Figure 4A. Given the presence of a closed loop, an inconsistency test was performed (Figure S7), and the results revealed no significant inconsistency. A forest plot (Figure 4B) showed that both the proximal AV and SV were associated with a substantially lowered risk of catheter occlusion compared with the distal AV. Among these tip locations, the SV was associated with the lowest incidence of catheter occlusion (RR =0.12; 95% CrI: 0.05–0.27; SUCRA =92.2%), followed by the proximal AV (RR =0.32; 95% CrI: 0.13–0.72; SUCRA =50.6%). The pairwise comparisons are detailed in Table S7.

Figure 4 BNMA of catheter occlusion. (A) Network diagram. (B) Forest plot of relative effects. BNMA, Bayesian network meta-analysis; CrI, credible interval; SUCRA, surface under the cumulative ranking curve.

Phlebitis

Seven studies provided data on phlebitis, involving a total of 960 participants and reporting 22 phlebitis events. The network plot is shown in Figure 5A. Given the presence of a closed loop, an inconsistency test was carried out, and the results (Figure S8) indicated no significant inconsistency. A forest plot (Figure 5B) demonstrated that in comparison to the distal AV, the SV was associated with a statistically significant reduction in the incidence of phlebitis(RR = 0.31; 95% CrI: 0.09–0.94; SUCRA =83.9%). Although the CrI approached the threshold for statistical significance, the effect estimate (RR) and ranking probability (SUCRA) consistently indicated that SV ranked among the better-performing locations. Pairwise comparisons are provided in Table S7.

Figure 5 BNMA of phlebitis. (A) Network diagram. (B) Forest plot of relative effects. BNMA, Bayesian network meta-analysis; CrI, credible interval; SUCRA, surface under the cumulative ranking curve.

Subgroup analysis

Results showed that SV catheterization significantly reduced the incidence of mechanical phlebitis in MC placement compared with AV catheterization [RR =0.27; 95% confidence interval (CI): 0.08–0.94; P=0.04], with data pooled using a fixed-effects model (I²<50%; P=0.07). However, it did not significantly reduce unclassified phlebitis, with no statistical significance (RR =0.71; 95% CI: 0.18–2.79, P=0.63), and data were pooled via a fixed-effects model (I²<50%; P=0.77). Overall, SV catheterization significantly reduced the incidence of phlebitis in MC placement compared with AV catheterization (RR =0.41; 95% CI: 0.18–0.95; P=0.04), with pooling via a fixed-effects model (I²<50%; P=0.87). Details are shown in Figure S9.

Sensitivity analysis

To verify the robustness of the findings, we reanalyzed the data using a random-effects model in addition to the primary analysis based on the fixed-effect model. The results were consistent across both approaches, with SV showing a statistically significant reduction in the risk of overall complications, phlebitis, catheter occlusion, and CRT (Tables S5,S8). Furthermore, given that only two RCTs and seven cohort studies were included, we compared the combined analysis (RCT + cohort studies) with an analysis restricted to the seven cohort studies (Tables S9,S10). The two analyses yielded consistent results, supporting the robustness of the study conclusions.

Discussion

This BNMA included nine studies and compared the impact of different tip locations of MC on complications. Our findings demonstrated that tip placement in the SV can lower the incidence of overall complications, CRT, catheter occlusion, and phlebitis. Tip placement in the BV was associated with the lowest incidence of overall complications. In contrast, positioning the catheter tip at the axillo-subclavian was linked to markedly increased risk of CRT. In contrast, positioning the catheter tip at the ASVJ was associated with a statistically significant increase in the risk of CRT.

In controlling MC-related complications, this study compared four tip placement sites—distal AV, proximal AV, SV, and BV—and found that all four could effectively reduce the incidence of complications. In terms of effectiveness, the top three placements were BV, SV, and proximal AV. MCs are medium-length vascular access devices, typically inserted into upper arm veins such as the basilic vein, cephalic vein, or BrV, with the tip terminating in the axillary or SV but not extending beyond the BV (5). BV access is primarily indicated when the internal jugular vein lumen is small (e.g., in hypovolemic patients), in patients with a history of multiple catheterizations leading to limited alternative access routes, or when SV puncture is contraindicated (e.g., coagulopathy or anatomical abnormalities) (23). Ultrasound-guided puncture of the left BV in a lateral-to-medial direction is recommended to reduce the number of puncture attempts and mechanical complications (24). The SV, located behind the clavicle, is an important tributary of the superior vena cava and is suitable for critically ill patients requiring long-term venous access (e.g., hematologic malignancies) or when internal jugular vein puncture is contraindicated (e.g., neck trauma) (25). SV access techniques include the supraclavicular approach, with the puncture site above the clavicle via the supraclavicular fossa, and the infraclavicular approach, with the puncture site below the clavicle. In clinical practice, the infraclavicular route is more common, but a study has shown that the supraclavicular approach offers a shorter venipuncture time under ultrasound guidance (median 9 vs. 13 s for the infraclavicular route) and a lower incidence of mechanical complications (3.0% vs. 13.4%) (26). Proximal AV cannulation, which establishes venous access through the proximal AV, is primarily used for cardiac implantable electronic devices (CIEDs) such as pacemaker or defibrillator lead placement, as well as central venous catheter insertion. Compared with traditional SV puncture, AV puncture can avoid intrathoracic complications such as pneumothorax and hemothorax (27). When the catheter tip is positioned in the axillary or SV, complication rates such as infection and thrombosis are lower (4).

Regarding phlebitis, the network meta-analysis results showed that compared to PVs, tip placement in the SV was associated with a statistically significant reduction in the risk of phlebitis (RR =0.31; 95% CrI: 0.09–0.94). Subgroup analysis indicated that SV tip placement significantly reduced the risk of mechanical phlebitis compared with the AV (RR =0.27; 95% CI: 0.08–0.94), while the reduction in unclassified phlebitis did not reach statistical significance. This contrasts with Zhang et al.’s review (15), which found no significant difference between SV and AV. The discrepancy may stem from different outcome definitions: Zhang et al. pooled unclassified phlebitis, while our analysis distinguished subtypes and included one study using standardized Infusion Nurses Society criteria, which likely clarified SV’s protective effect. The anatomical stability of the SV—situated between the clavicle and first rib, supported by surrounding muscles and ligaments—reduces friction between the catheter and venous endothelium, thereby lowering inflammatory response risk (28,29). The larger lumen diameter of the SV (10–12 mm) compared to PVs (3–4 mm) also accommodates the catheter more easily, reducing turbulence and shear stress, which are key triggers of phlebitis (30,31). The non-significant effect on unclassified phlebitis may be due to heterogeneity in study definitions and the influence of chemical phlebitis risk factors (e.g., pH, osmolarity of infusates). Future studies should standardize phlebitis subtypes using INS (32) criteria and control for drug properties to more accurately isolate the anatomical effect of SV placement.

In terms of CRT, tip placement in the SV was associated with a markedly reduced risk compared with the distal AV (RR =0.34; 95% CrI: 0.13–0.92), while positioning at the ASVJ significantly increased CRT risk. This finding aligns with previous RCTs demonstrating that the SV’s high shear stress (18.6±4.2 dyn/cm²) may activate the KLF2 pathway, promoting thromboregulatory protein expression (26). The more acute angle of the distal AV promotes turbulent flow and platelet dwell time, increasing thrombosis risk (33). Anatomical gradient effects have been observed, where more central placement (e.g., SV) correlates with reduced thrombus formation (4). Ultrasound-guided placement into the SV or thoracic AV segment can further lower CRT risk by ensuring accurate tip position and minimizing endothelial injury from repeated punctures (34,35).

In terms of catheter occlusion, tip placement in the SV was associated with a statistically significant reduction in occlusion risk compared with PVs (RR =0.12; 95% CrI: 0.05–0.27). This protective effect may be related to the anatomical stability of the SV, which minimizes catheter displacement and mechanical friction. Evidence suggests that when the MC tip is positioned in the SV or the thoracic segment of the AV, catheter-related complications—including occlusion—are significantly reduced (4). By contrast, Zhang et al. did not observe a significant difference in catheter obstruction between SV and AV placements (15). This inconsistency may be explained by differences in study inclusion: Zhang et al. included eight studies, one of which could not be traced, whereas our analysis incorporated nine studies, including two additional trials beyond those reported by Zhang. These differences in evidence sources may have contributed to the discrepancy in outcomes. The left SV route, due to its favorable symmetry with the BV, has a lower malposition rate than the right (4.5% vs. 13.8%), particularly in avoiding misplacement into the ipsilateral internal jugular vein (34). Furthermore, ultrasound-guided insertion, especially via the right SV, can reduce malposition from 7% to 2.9% and lower overall complication rates, including occlusion (36). Therefore, for patients requiring long-term catheterization, SV should be prioritized (grade 1 recommendation).

Our findings support the clinical feasibility and safety of SV tip placement for MCs, particularly under real-time ultrasound guidance (26,37). Both supraclavicular and infraclavicular approaches are viable, with the supraclavicular route offering shorter puncture times (median 9 vs. 13 s) and comparable complication rates (26). Ultrasound guidance improves first-attempt success rates by 32%, reduces arterial puncture (from 17% to 6%) and pneumothorax risk, and allows real-time confirmation of guidewire position to lower malposition rates (27,36). Standardized operator training—including simulation-based practice and ultrasound navigation skills—is essential to minimize mechanical complications (38). For pediatric patients, ultrasound-guided supraclavicular BV access is recommended, with adjustments in catheter size and insertion side to match developmental anatomy (24,39).

Future research should prioritize multicenter, high-quality RCTs using standardized protocols (40), uniform tip confirmation methods (e.g., ultrasound plus radiography), and consistent definitions for mechanical phlebitis, chemical phlebitis, and thrombosis (35,41). Studies in high-risk groups (e.g., oncology, pediatric) are needed, alongside consensus development of core outcome sets. Mechanistic studies using 4D flow MRI and biomarker monitoring (e.g., IL-6, P-selectin) could clarify hemodynamic and endothelial responses (42,43). Economic evaluations and virtual reality-based training systems may further support the adoption of optimized, evidence-based SV placement strategies in clinical practice.

This study is the first BNMA to systematically evaluate the impact of the tip location of MCs on several complications, including phlebitis, thrombosis, and occlusion. Our study addresses certain deficiencies in research on MC, as previous studies have focused primarily on PICCs. In addition, this study is the first to quantify the anatomical gradient in complication risk from the peripheral to central veins. Our results confirmed the protective effect of the SV across multiple complications (such as a 25-fold reduction in thrombosis risk and a 69% reduction in phlebitis risk), thus providing evidence-based insights for clinical practice.

Several limitations should be considered when interpreting our findings. First, only two RCTs were included, both of which were of relatively low quality, and several studies were not available in full text, preventing a complete assessment of their methodological rigor. Furthermore, most studies originated from a single country, which may limit the generalizability of the results to broader clinical settings. Second, some outcomes had wide 95% CrIs, likely due to small sample sizes or extreme values, warranting cautious interpretation in clinical practice. Third, tip confirmation methods (e.g., surface landmark positioning, ultrasound guidance, or X-ray confirmation) were not standardized across studies, which may have introduced residual confounding. At the same time, the protective effect of BV placement may have been underestimated due to the limited number of relevant studies. Moreover, the lack of long-term follow-up data (>30 days) precluded evaluation of the sustained protective effect of SV placement. In addition, several included studies did not systematically report infusion fluid or drug characteristics (e.g., pH, corrosiveness, and type), and variations in infusion regimens could confound the association between tip location and complication risk. Finally, although we performed phlebitis subtype analyses, the significant effect observed for mechanical phlebitis was based on limited studies, and unclassified phlebitis outcomes may have been influenced by factors such as drug properties. There was also notable heterogeneity in the definitions and diagnostic criteria for phlebitis, particularly for deep vein placements (SV/BV) where specific symptoms are lacking; in some cases, superficial phlebitis criteria (e.g., erythema, palpable cord) were applied, potentially leading to underdiagnosis. Although sensitivity analyses excluding studies without imaging-based diagnosis confirmed the stability of SV’s advantage, variability in diagnostic standards remains a key limitation. Therefore, future studies should adopt standardized diagnostic criteria for phlebitis and thrombosis (e.g., combining imaging and inflammatory markers), extend follow-up durations, and use uniform insertion and maintenance protocols to strengthen the robustness and generalizability of the findings.

Conclusions

This study demonstrated that positioning the tip of MCs in the SV may be associated with a statistically significant reduction in the risk of phlebitis (RR =0.31), CRT (RR =0.04), and catheter occlusion (RR =0.48). The protective effect of SV may be related to its favorable hemodynamic and anatomical positions. Future research should focus on conducting RCTs to standardize catheter tip confirmation techniques (ultrasound guidance is recommended as the preferred option), assessing the safety of long-term catheterization (>30 days).

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the PRISMA NMA reporting checklist. Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-333/rc

Peer Review File: Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-333/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-333/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bundgaard Madsen E, Sloth E, Skov Illum B, et al. The clinical performance of midline catheters-An observational study. Acta Anaesthesiol Scand 2020;64:394-9. [Crossref] [PubMed]
2. Bentridi A, Giroux MF, Soulez G, et al. Midline Venous Catheter vs Peripherally Inserted Central Catheter for Intravenous Therapy: A Randomized Clinical Trial. JAMA Netw Open 2025;8:e251258. [Crossref] [PubMed]
3. Qin KR, Ensor N, Barnes R, et al. Standard Versus Long Peripheral Catheters for Multiday IV Therapy: A Randomized Controlled Trial. Pediatrics 2021;147:e2020000877. [Crossref] [PubMed]
4. Zhao L, Fan X, Zhao L, et al. Midline catheter tip position and catheter-related complications in antimicrobial therapy: A multi-center randomized controlled trial. Int J Nurs Stud 2023;141:104476. [Crossref] [PubMed]
5. Tripathi S, Kumar S, Kaushik S. The Practice and Complications of Midline Catheters: A Systematic Review. Crit Care Med 2021;49:e140-50. [Crossref] [PubMed]
6. Seo H, Altshuler D, Dubrovskaya Y, et al. The Safety of Midline Catheters for Intravenous Therapy at a Large Academic Medical Center. Ann Pharmacother 2020;54:232-8. [Crossref] [PubMed]
7. Cox C, Roberts LN. Basics of diagnosis and treatment of venous thromboembolism. J Thromb Haemost 2025;23:1185-202. [Crossref] [PubMed]
8. Li X, Zhang H, Li H, et al. Comparison of complications and indwelling time in midline catheters versus central venous catheters: A systematic review and meta-analysis. Int J Nurs Pract 2024;30:e13301. [Crossref] [PubMed]
9. Paje D, Walzl E, Heath M, et al. Midline vs Peripherally Inserted Central Catheter for Outpatient Parenteral Antimicrobial Therapy. JAMA Intern Med 2025;185:83-91. [Crossref] [PubMed]
10. Hassani M, Mohebbi M, Tavallaei M, et al. May position of hemodialysis catheter tip have a direct effect on its patency? Positive results of a preliminary study on its rotation. Eur J Transl Myol 2022;32:10537. [Crossref] [PubMed]
11. Wu HW, Qiu YR, Chen LY. Midline Catheter Tip Position and Its Association with Complications. Lingnan Journal of Emergency Medicine 2020;25:550-1.
12. Schechter MS, Baron SW, Gohari A, et al. Midline Catheter-Associated Thrombosis (MCAT): Does Tip Location in the Axillary Vein Increase Risk? J Infus Nurs 2024;47:363-8. [Crossref] [PubMed]
13. Elli S, Pittiruti M, Pigozzo V, et al. Ultrasound-guided tip location of midline catheters. J Vasc Access 2020;21:764-8.
14. Zhao LF, Cai ZY, Fan XP, et al. Application comparison of different tip positions of midline catheters. Chinese Journal of Nursing 2022;57:517-24.
15. Zhang H, Li X, Sun W, et al. Comparison of complications and indwelling time of midline catheter at different tip locations: A systematic review and meta-analysis. J Vasc Access 2024;25:1757-66.
16. Zhai Y, Zhang Y, Gou H. Pharmacological treatment of pediatric nocturnal enuresis: a systematic review and network meta-analysis. Pediatr Nephrol 2025;40:2795-806. [Crossref] [PubMed]
17. Nikolakopoulou A, Higgins JPT, Papakonstantinou T, et al. CINeMA: An approach for assessing confidence in the results of a network meta-analysis. PLoS Med 2020;17:e1003082. [Crossref] [PubMed]
18. Liu YF, Li CH. A comparative study of midline catheters with different tip positions in the perioperative period in patients with gastrointestinal tumors. Chinese Journal of Cancer Prevention and Treatment 2019;26:211, 213.
19. Bai J, Guo Y. Comparison of mid-length catheter tip position in the anterior axillary line and midclavicular line. Electronic Journal of General Stomatology 2019;6:133, 137.
20. Lang L, Lu HP, Xiao QG, et al. Application of different midline catheters tip location in patients with hepatobiliary surgery related diseases. Clinical Research and Practice 2020;5:149-51.
21. Bai J, Guo Y, Mo YZ, et al. Effects of different tip positions of midline catheter in elderly patients. Practical Geriatrics 2022;36:1300-3.
22. Fu J, Zhao L, Tian Y, et al. Impact of terminal tip location of midline catheters on catheter-related complications and indwelling duration in elderly patients. Nurs Open 2023;10:2349-56. [Crossref] [PubMed]
23. Kumar A, Sinha C, Kumar A, et al. Comparison between in-plane and out-of-plane techniques for ultrasound guided cannulation of the left brachiocephalic vein in pediatric population: A randomised controlled trial. Anaesth Crit Care Pain Med 2023;42:101247. [Crossref] [PubMed]
24. Timsit JF, Baleine J, Bernard L, et al. Expert consensus-based clinical practice guidelines management of intravascular catheters in the intensive care unit. Ann Intensive Care 2020;10:118. [Crossref] [PubMed]
25. Heidenreich D, Hansen E, Kreil S, et al. The insertion site is the main risk factor for central venous catheter-related complications in patients with hematologic malignancies. Am J Hematol 2022;97:303-10. [Crossref] [PubMed]
26. Kim YJ, Ma S, Yoon HK, et al. Supraclavicular versus infraclavicular approach for ultrasound-guided right subclavian venous catheterisation: a randomised controlled non-inferiority trial. Anaesthesia 2022;77:59-65. [Crossref] [PubMed]
27. Vitali F, Malagù M, Bianchi N, et al. Ultrasound-Guided Venous Axillary Access Versus Standard Fluoroscopic Technique for Cardiac Lead Implantation: ZEROFLUOROAXI Randomized Trial. JACC Clin Electrophysiol 2024;10:554-65. [Crossref] [PubMed]
28. Su J, Kusumoto FM, Zhou X, et al. How to perform extrathoracic venous access for cardiac implantable electronic device placement: Detailed description of techniques. Heart Rhythm 2022;19:1184-91. [Crossref] [PubMed]
29. Zhou M, Dong S, Zhang J, et al. Effects of the low-speed continuous infusion catheter technique on double-lumen central venous catheters: A randomized controlled trial. Int J Nurs Stud 2024;151:104676. [Crossref] [PubMed]
30. Yakusheva AA, Butov KR, Bykov GA, et al. Traumatic vessel injuries initiating hemostasis generate high shear conditions. Blood Adv 2022;6:4834-46. [Crossref] [PubMed]
31. Dobrescu A, Constantin AM, Pinte L, et al. Effectiveness and safety of methods to prevent bloodstream and other infections and noninfectious complications associated with peripherally inserted central catheters: A systematic review and meta-analysis. Clin Infect Dis 2025;ciaf063. [Crossref] [PubMed]
32. Gorski LA, Hadaway L, Hagle ME, et al. Infusion Therapy Standards of Practice, 8th Edition. J Infus Nurs 2021;44:S1-S224.
33. Yuen HLA, Tran H, Chunilal S. Upper Extremity Deep Vein Thrombosis: Current Knowledge and Future Directions. Semin Thromb Hemost 2021;47:677-91. [Crossref] [PubMed]
34. Shin KW, Park S, Jo WY, et al. Comparison of Catheter Malposition Between Left and Right Ultrasound-Guided Infraclavicular Subclavian Venous Catheterizations: A Randomized Controlled Trial. Crit Care Med 2024;52:1557-66. [Crossref] [PubMed]
35. Boulet N, Muller L, Rickard CM, et al. How to improve the efficiency and the safety of real-time ultrasound-guided central venous catheterization in 2023: a narrative review. Ann Intensive Care 2023;13:46. [Crossref] [PubMed]
36. Adrian M, Kander T, Lundén R, et al. The right supraclavicular fossa ultrasound view for correct catheter tip positioning in right subclavian vein catheterisation: a prospective observational study. Anaesthesia 2022;77:66-72. [Crossref] [PubMed]
37. Zawadka M, La Via L, Wong A, et al. Real-Time Ultrasound Guidance as Compared With Landmark Technique for Subclavian Central Venous Cannulation: A Systematic Review and Meta-Analysis With Trial Sequential Analysis. Crit Care Med 2023;51:642-52. [Crossref] [PubMed]
38. Adrian M, Borgquist O, Kröger T, et al. Mechanical complications after central venous catheterisation in the ultrasound-guided era: a prospective multicentre cohort study. Br J Anaesth 2022;129:843-50. [Crossref] [PubMed]
39. Singh NS, Grimes J, Gregg GK, et al. “Save the Vein” Initiative in Children With CKD: A Quality Improvement Study. Am J Kidney Dis 2021;78:96-102.e1. [Crossref] [PubMed]
40. Annetta MG, Spencer TR, Pittiruti M. The never-ending confusion between subclavian vein and axillary vein. J Vasc Access 2025. [Epub ahead of print]. doi: 10.1177/11297298251352477.
41. Boulet N, Pensier J, Occean BV, et al. Central venous catheter-related infections: a systematic review, meta-analysis, trial sequential analysis and meta-regression comparing ultrasound guidance and landmark technique for insertion. Crit Care 2024;28:378. [Crossref] [PubMed]
42. Villa A, Hermand V, Bonny V, et al. Improvement of central vein ultrasound-guided puncture success using a homemade needle guide-a simulation study. Crit Care 2023;27:379. [Crossref] [PubMed]
43. Faccini B, Aula F, Fontana V, et al. Comparing results of midline catheter and peripherally inserted central catheter in cancer patients under chemotherapy. J Vasc Access 2024. [Epub ahead of print]. doi: 10.1177/11297298241289326.

(English Language Editor: J. Gray)