Implantable Ports vs Peripherally Inserted Central Catheters in Breast Cancer Chemotherapy: A Comprehensive Meta-Analysis

Introduction

Breast cancer remains one of the most prevalent malignancies worldwide, accounting for a significant proportion of cancer diagnoses and deaths among women [1]. Its risk factors include, hormonal influences, lifestyle factors, advancing age, and genetic predisposition, such as BRCA mutations [2]. Early detection through mammography and improved treatment strategies have substantially enhanced survival rates, yet advanced or aggressive cases often require multimodal therapies [3–6]. Treatment approaches vary based on tumor biology and stage and encompass surgery, radiation, targeted therapies, hormonal agents, and chemotherapy [7,8].

Chemotherapy plays a critical role in breast cancer management. It is the primary treatment for metastatic disease, aiming to prolong survival and alleviate symptoms. Administering chemotherapy drugs requires reliable venous access, due to the need for repeated infusions over weeks or months [9,10]. This necessity has driven the development of long-term vascular access devices, such as peripherally inserted central catheters (PICC) and implantable port catheters (IPC), which minimize complications associated with frequent peripheral intravenous access. However, the selection between these devices remains debated in clinical practice and is influenced by divergent safety profiles, procedural demands, and patient-centered outcomes [11,12]. IPCs, which are surgically embedded beneath the skin, enable direct central venous delivery of chemotherapy and demonstrate reduced catheter-related infections, owing to their protected placement and minimized exposure to external contaminants [13–15]. In contrast, PICCs – inserted via peripheral veins in the arm – offer rapid, nonsurgical placement, which is often favored for short-term needs or patients unsuitable for surgery. However, PICCs carry elevated risks of bloodstream infections and venous thrombo-embolism, especially among those receiving thrombogenic therapies or with coagulation disorders. While PICCs may align with cost-effective workflows, because of simpler insertion, IPCs are linked to enhanced patient comfort and adherence, as their fully implanted design reduces visibility, requires less maintenance, and avoids the psychosocial burden of external catheters. Despite these trade-offs, institutional preferences and logistical factors often influence clinical decision [1,2,16,17].

Although IPC and PICC differ in methods of placement, both serve as central venous access devices for chemotherapy in patients with breast cancer. We include them as comparators because of their shared use in clinical practice for chemotherapy administration and aim to assess the complication risks between the two. This meta-analysis synthesizes comparative evidence on complications, efficacy, and quality of life, aiming to clarify optimal strategies for venous access in chemotherapy-treated breast cancer patients.

Material and Methods

SEARCH STRATEGY:

A comprehensive literature search was conducted across 4 databases – PubMed, EMBASE, Cochrane Library, and ScienceDirect – to identify randomized controlled trials comparing IPCs and PICCs in patients with breast cancer. The search encompassed studies published from Jan 1, 2000, to March 15, 2025, using Medical Subject Headings (MeSH) terms such as “breast neoplasms”, “chemotherapy”, “implantable venous port”, and “peripherally inserted central venous catheter” (Table 1). Supplementary articles were identified by manually screening reference lists of retrieved publications. We restricted the language of the literature to only English. A total of 524 articles were found. Fifteen were related to PICCs and IPCs, of which, 5 articles were selected. This meta-analysis is registered on PROSPERO website (registration ID: CRD420251028346).

SELECTION CRITERIA:

The inclusion criteria were as follows: inclusion of patients with breast cancer who were prescribed chemotherapy; direct comparison of IPC and PICC; data reported on catheter-related complications, such as thrombosis and infection, chemotherapy adherence, including delays and discontinuations, and cost-effectiveness.

The exclusion criteria were studies lacking primary data, such as meta-analyses, conference abstracts, case reports, and narrative reviews. Duplicate publications from the same trial were included only if they reported distinct outcomes.

DATA EXTRACTION:

Two reviewers independently extracted data on study characteristics (author, publication year, country); patient demographics (age, cancer stage); complication rates (thrombosis, infection subtypes); and chemotherapy interruptions. The inter-rater agreement was calculated using Cohen’s kappa statistic. Disagreements between reviewers were resolved through consultation with a third reviewer.

OUTCOME ASSESSMENT:

Complication rates were defined as the incidence per patient of catheter-related complications, including thrombosis (specifically deep vein thrombosis and catheter-related thrombosis), and infection (localized or systemic) stratified by type and frequency. Operative time was defined as the duration from catheter insertion to wound closure or catheter fixation completion.

QUALITY APPRAISAL:

Randomized controlled trial quality was assessed using the Jadad scale, which has a total score ranging from 0 to 7 points and evaluates randomization rigor (0–2 points), blinding (0–2 points), and attrition management (0–2 points). Studies scoring ≥4 points were classified as “high” quality, while studies scoring below this threshold were classified as “low” quality. The ROBINS-I tool was used to evaluate the risk of bias in non-randomized studies, including cohort and case-control studies. The GRADE system was used to assess the overall quality of evidence across studies. For each study, we evaluated risk of bias, inconsistency, indirectness, imprecision, and publication bias.

STATISTICAL ANALYSIS:

Analyses were performed using R programming (Version 4.5.0) software. The “meta” package was applied. Dichotomous variables with risk ratios (RR) were reported along with 95% confidence intervals (CI). Heterogeneity was quantified using I² statistics and χ² tests. Publication bias was assessed via funnel plots, and Egger’s and Begg’s tests were performed. To assess the robustness of the pooled effect estimates, a leave-one-out sensitivity analysis was conducted. Statistical significance was set at P<0.05.

Results

Figure 1 shows the selection process. The literature search and study selection process followed the PRISMA guidelines. A flowchart detailing the identification, screening, and inclusion stages is shown in Figure 1. A comprehensive search was conducted across electronic databases and clinical trial registries, yielding 524 initial records. A total of 168 duplicate records were removed. Titles and abstracts of 356 records were screened for relevance. Of these, 76 records were retained for full-text retrieval and eligibility assessment. The remaining 280 records were excluded due to irrelevance to the research question (eg, lack of comparison between peripherally inserted central catheters and intravenous catheters). Full-text evaluation of the 76 retrieved reports led to further exclusions: 39 studies were excluded for not comparing PICC vs IPC, and 32 studies were excluded for insufficient data or inability to extract outcomes. Ultimately, 5 studies met all predefined inclusion criteria and were included in the meta-analysis.

Table 2 shows the overview of the study characteristics. The studies were conducted in Italy, Sweden, France, and Canada. All studies compared PICC and IPC, primarily for chemotherapy delivery in patients with breast cancer. Each study included between 56 and 448 patients. Breast cancer treatments included neoadjuvant therapy (Pinedili et al, 2024; Clemons et al, 2020), adjuvant therapy (Lefebvre et al, 2015; Clatot et al, 2020), and mixed or unspecified approaches (Utas et al, 2025). Study designs included randomized controlled trials (Clatot et al, 2020; Clemons et al, 2020), observational studies (Pinedili et al, 2024; Lefebvre et al, 2015 [retrospective]), and a post-hoc analysis (Utas et al, 2025). The quality of the included studies was assessed using the ROBINS-I, Jadad scale, and GRADE scores. The ROBINS-I indicated varying risks of bias, with the observational studies showing higher risks (eg, “serious” for Pinelli et al, 2024). Jadad scale scores for randomized controlled trials ranged from 6/8 to 7/8, reflecting generally high methodological quality. GRADE ratings were primarily low to moderate.

Figure 2 displays the forest plot, which synthesized data from 5 studies (n=1125 participants) comparing complication risks between PICCs and IPCs in patients with breast cancer. Individual study results were as follows: (1) Pinelli et al (2024) reported an RR of 1.214 (95% CI: 0.470–3.140), with 9 events out of 108 in the PICC group and 7 out of 102 in the IPC group, indicating no significant difference in complications [18]. (2) Utas et al (2025) reported an RR of 1.465 (95% CI: 0.798–2.690), with 20 events out of 78 in the PICC group and 14 out of 80 in the IPC group, showing a non-significant trend toward higher risk with PICC [19]. (3) Lefebvre et al (2015) reported an RR of 2.541 (95% CI: 1.279–5.049), with 18 events out of 158 in the PICC group and 13 out of 290 in the IPC group, indicating a statistically significant higher risk with PICC [20]. (4) Clatot et al (2020) reported an RR of 2.117 (95% CI: 1.039–4.312), with 21 events out of 126 in the PICC group and 10 out of 127 in the IPC group, indicating a significant 2-fold higher risk with PICC [21]. (5) Clemons et al (2020) reported an RR of 1.164 (95% CI: 0.349–3.888), with 5 events out of 29 in the PICC group and 4 out of 27 in the IPC group, showing a non-significant difference, likely due to the small sample size [22].

The pooled analysis showed that the overall risk ratio was 1.750 (95% CI: 1.244–2.461), indicating a statistically significant increased risk of complications with PICC, compared with IPC. The heterogeneity among studies was low, with I2=20%. Egger’s test (P=0.411) and Begg’s test (P=0.624) yielded non-significant P values, indicating no strong evidence of publication bias in the meta-analysis. This aligns with the lack of obvious asymmetry in the funnel plot (Figure 3). A leave-one-out sensitivity analysis was performed to evaluate the robustness of the pooled effect estimate by sequentially omitting each included study. The overall pooled RR for complications associated with PICC compared with IPC remained statistically significant across all studies, with RR values ranging from 1.563 (95% CI: 1.057–2.312) to 1.902 (95% CI: 1.266–2.858). No single study substantially altered the direction or magnitude of the effect, suggesting high stability of the results and minimal influence of outlier studies (Figure 4).

Discussion

This meta-analysis of 1125 patients with breast cancer across 5 studies demonstrated a clinically significant increased risk of complications with PICCs, compared with IPCs. The pooled RR of 1.75 (95% CI: 1.24–2.46) underscores the superior safety profile of IPCs, particularly in reducing thromboembolic and infectious complications. These findings align with those of prior observational studies and meta-analyses in mixed oncology populations, which have similarly highlighted the heightened thrombogenicity and infection susceptibility of PICCs, due to their peripheral insertion site and prolonged external exposure [23–26]. However, our study strengthens the evidence by focusing specifically on patients with breast cancer – a population often exposed to thrombogenic therapies (eg, anthracyclines) and hormonal agents that can amplify catheter-related risks. Notably, the 2 largest studies (Lefebvre et al, 2015; Clatot et al, 2020) demonstrated statistically significant risks favoring IPC, while smaller trials reported non-significant trends in the same direction. This pattern suggests that larger sample sizes are critical for detecting true differences, as smaller studies can lack power to resolve moderate effect sizes. The elevated thrombosis risk with PICCs likely stems from their narrower-caliber peripheral veins, which are prone to endothelial injury and stasis during chemotherapy. Similarly, the higher infection rates may reflect greater microbial colonization at the external entry sites of PICCs, compared with the subcutaneously embedded ports of IPCs [27–29]. These mechanistic insights, coupled with our pooled results, support guidelines advocating IPC use in patients requiring prolonged chemotherapy. Beyond safety, our analysis has implications for treatment adherence and cost-effectiveness. While PICCs offer rapid, nonsurgical placement, their association with complications can indirectly increase chemotherapy interruptions – a factor not uniformly quantified in included studies but critical to oncologic outcomes. IPCs, although they require minor surgery, reduce maintenance burden and visibility, potentially improving patient satisfaction and adherence. Economically, while PICCs may appear cost-effective upfront, their higher complication rates likely escalate long-term expenditures through added imaging, anticoagulation, and infection management.

Singh et al reported higher patient satisfaction with IPCs despite socioeconomic disparities in access preference [30]. A prospective study of 311 breast cancer patients with PICC lines found a 51.4% thrombosis rate, with most cases occurring within 2 weeks of insertion. Reduced limb activity and obesity were identified as independent risk factors [31]. Clinical strategies should prioritize managing obesity, monitoring coagulation, and ensuring proper catheter placement to mitigate thrombosis risks [9]. Xu et al reported a 4.62% PICC-related thrombosis rate in 780 patients with breast cancer, with half of cases occurring within 30 days of insertion [32]. Robinson et al systematically reviewed 15 studies comparing vascular access strategies in patients with early-stage breast cancer receiving chemotherapy. They found lower median complication rates with IPCs than with PICCs [33]. Su et al analyzed 4954 breast cancer patients undergoing totally implantable venous access device removal and identified low complication rates, including bleeding, removal difficulty, and delayed wound healing. Longer indwelling time correlated with higher catheter rupture/fracture risks. They underscored the need for timely removal and material considerations to optimize outcomes [34]. In a cross-sectional study of 223 breast cancer patients with PICC chemotherapy, Zhang et al demonstrated that illness uncertainty directly predicts cancer-related fatigue, highlighting the need for interventions targeting psychological resilience and self-care training to mitigate cancer-related fatigue [35]. Liu et al identified a 21.6% PICC-related thrombosis rate in 2227 patients with cancer, with key risk factors including age >65 years, male sex, active infection, polyurethane catheters, and cisplatin use. They advocate prioritizing silicone catheters, infection control, and intensified monitoring in high-risk subgroups (eg, gastrointestinal cancers, prolonged dwell time) to mitigate thrombosis [36].

In our study, several limitations warrant consideration. First, the inclusion of randomized controlled trials and observational studies introduces potential confounding, although relatively low heterogeneity mitigates concerns about design-related bias. Clinical heterogeneity across studies might be due to differences in patient populations and healthcare settings. Second, key demographic data, such as mean age and thrombotic risk factors (eg, BRCA status, obesity), were inconsistently reported, and the number of included studies is relatively small, limiting the statistical power to perform meaningful subgroup analyses or meta-regression to fully explore potential sources of heterogeneity. Therefore, the robustness of our results in different clinical contexts or patient subgroups could not be fully assessed. Third, temporal and geographical variations in catheter insertion protocols, such as anticoagulant use and aseptic techniques, can influence complication rates but were not explored. Finally, patient-reported outcomes, including pain and body image concerns, were scarcely reported, limiting insights into quality-of-life trade-offs. In the future, prospective randomized controlled trials with standardized outcome measures are needed to confirm these findings and address gaps in adherence and cost data. We should also explore risk stratification through these methods: identifying subgroups, such as HER2-positive patients on dual antiplatelet therapy, who derive maximal benefit from IPC; assessing safety of antimicrobial-impregnated PICCs or ultrasound-guided IPC placement in resource-limited settings; and evaluating late complications, including port-site metastases and chronic venous stenosis, beyond chemotherapy cycles.

Conclusions

In patients with breast cancer receiving chemotherapy, IPCs are associated with fewer complications than PICCs, supporting their preferential use in non-short-term treatment and in patients at high risk of thrombosis or infection. These findings support updating clinical guidelines to optimize vascular access strategies in breast cancer care.