Vascular access devices (VADs) are essential components in the management of hospitalized patients, with approximately 90% of inpatients requiring a VAD for the safe administration of medications, fluids, electrolytes, parenteral nutrition, and blood products (1). Even in outpatient settings, VADs play a crucial role in ensuring prolonged therapies such as chemotherapy and antibiotic therapy. However, they are not free from complications (2).

VAD-associated complications are various, including mechanical complications at insertion, thrombosis, infection, phlebitis, occlusion, and dislocation, negatively impacting outcomes and increasing the workload for healthcare staff (3). Complications arise not only from an improper VAD insertion and management but also from an inappropriate VAD selection. Selecting the most appropriate VAD for the intended use is then fundamental to ensuring optimal patient care, improving safety, and reducing costs (4, 5). Clinically, various strategies have been implemented to enhance the appropriateness of VAD, including algorithms of choice, hospital policies, and, obviously, training (6). Nevertheless, the actual application of these strategies remains lacking, as does the evaluation of VAD appropriateness in the real world.

The present study aimed to evaluate the appropriateness of VAD selection in the medical wards of a tertiary care hospital.

In this study, VAD appropriateness was defined as the correctness of device selection based on the patient’s therapeutic needs, as recommended by two of the most renown international reference documents: the ERPIUP Consensus (World Congress of Vascular Access - WoCoVA) and the Infusion Therapy Standards of Practice (Infusion Nursing Society - INS) (7, 8).

Based on these documents, it was considered inappropriate:

CVAD’s intraprocedural complications and PVAD’s failure rates were also collected.

Eligible patients were those admitted to the participating medical wards with an indication for a vascular access device (VAD). Patients below 18 years of age were excluded from the study.

In the authors’ hospital, policies on the selection and use of VADs have been in place since 2014 and updated every five years.

VADs were defined according to the terminology recently proposed by the World Congress of Vascular Access (WoCoVA) – the NAVIGATE project(9) – which defined central vascular access devices (CVADs) as centrally inserted central catheter (CICC), peripherally inserted central catheter (PICC), femorally inserted central catheter (FICC), port and tunnelled-cuffed catheter (Tc); and peripheral vascular access devices (PVADs) as short peripheral cannulas (SPC), long peripheral cannulas (LPC), and midline catheter (MC).

Data were collected using REDCap® (Research Electronic Data Capture), a web application developed by Vanderbilt University for building and managing online surveys and databases. Patients’ demographics, type of VAD, reason for insertion, duration, complications, and adherence to best practice guidelines were recorded.

Patients were enrolled at the time of device insertion, as determined independently by the prescribing physician, in accordance with the clinical practice of the medical ward. Before enrolment, informed consent was obtained from all patients. The observation period ended when the patient was discharged from the ward. No further follow-up assessments were scheduled beyond this point.

A total of 23 out of 322 VADs (7.1%) were CVADs. Seven PICCs were inserted, accounting for 30.4% of all CVADs, with a median dwell time of 6.5 days (IQR 5–8). Four CICCs were inserted, representing 17.4% of CVADs, with a median dwell time of 7 days (IQR 4–9). Three FICC (13% of CVADs) were inserted, with a median dwell time of 12 days (IQR 10–17.5). Six ports (26.2%) were implanted.

Among CVADs, 82.6% were appropriate and 17.4% were inappropriate. One CICC was appropriate (25%) and 3 inappropriate (75%). PICC and FICC showed the highest appropriateness rates (n = 10; 100%; and n = 3; 100% respectively). Five ports (83.3%) were appropriate, while 1 (16.7%) was inappropriate. A detailed representation of CVAD’s appropriateness is shown in Table 2 and Fig. 3.

Clinical indications for CVAD included chemotherapy (n = 14; 60.9%), fluid therapy (n = 10; 43.5%), pain management (n = 10; 43.5%), antibiotic therapy (n = 5; 21.7%), transfusions (n = 6; 26.1%), contrast medium for diagnostic procedures (n = 3; 13.0%), total parenteral nutrition (n = 2; 8.7%), and vasoactive amine administration (n = 2; 8.7%).

A total of 299 out of 322 VADs (92.8%) were PVADs, out of which 129 (43.1%) were appropriate; conversely, 170 (56.9%) were inappropriate. Out of 299 PVADs, 257 (85.9%) were short peripheral catheters (SPC), 32 (10.7%) were long peripheral catheters (LPC), and 10 (3.3%) were midline catheters (MC).

MC showed the highest appropriateness rate (70%), followed by LPC (56.2%). SPC had the lowest appropriateness rate (40.5%), with a total of 170 PVADs (56.85%) deemed as inappropriate (Table 2 and Fig. 3).

The majority of PVADs were used for antibiotic therapy (191 cases, 63.9%) and pain management (193 cases, 64.5%). Another indication was fluid therapy, accounting for 169 cases (56.5%), representing more than half of the inserted PVADs. Other indications included the administration of blood products (n = 28; 9.4%) and contrast medium infusion (n = 19; 6.4%).

Median dwell time of PVADs was 4 days for SPC, 6 days for LPC and 5 days for MC.

This study explored the appropriateness of vascular access selection in two wards of a tertiary care University Hospital. Despite the policies for appropriate device choice having been implemented for many years, in this study, a high rate of inappropriate VAD selection was found, occurring in 17.4% and 57.8% of CVADs and PVADs, respectively. The lower rate of inappropriateness among CVADs may reflect a greater awareness of the increased risks of complications associated with CVADs compared to PVADs, including arterial or lung puncture, thrombosis, and infection (8, 10). Therefore, greater caution in their indication and management may have been adopted. Moreover, the administration of solutions not fully compatible with the peripheral route via a PVAD is a major factor contributing to inappropriateness and may have influenced the results.

Interestingly, among CVADs, a notable difference in terms of appropriateness was found. PICC and FICC showed a 100% appropriateness rate, whereas CICC showed only 25%. This finding cannot be easily explained. It may be that CICC is more frequently placed in emergencies and then left in place for more than 48 hours and used for drugs compatible with the peripheral route. Additionally, not all the CICCs in the authors’ hospital are suitable for contrast media infusion, and they may have been misused. Moreover, in the authors’ hospital, PICC and FICC are implanted by a Vascular Access Team (VAT). VAT is a multidisciplinary group of vascular access specialists (11) whose duties include not only inserting a VAD but also verifying their indication and selecting the most appropriate one for that specific patient: this may have contributed to a greater appropriateness. On the contrary, CICCs are often implanted by “occasional inserters” (i.e., non-vascular access specialists), who may also have been responsible for the greater rate of intraprocedural complications (i.e., multiple venipunctures) found among the CICC group compared to PICC and FICC ones. One PORT out of 6 was considered inappropriate, likely due to its use while other PVADs were available for the infusion of peripherally compatible solutions for more than 48 hours. Notably, PICC was the most represented CVAD (43.5%), well beyond CICC (17.4%) and FICC (13%). According to local hospital policies and consensus documents (12, 13), PICC is indeed the CVAD of choice in the hospitalised patient. This is easily explained by several reasons, including: the lack of major risks at the insertion even in fragile patients, such as those with coagulation disorders or cardiorespiratory problems (14), low risk of late complications (13, 15, 16), and the possibility of discharging the patient at long-term facilities or at home with the PICC.

Among PVADs, SPC and LPC showed the greatest inappropriateness rate, being 59.5% and 43.8% respectively. Possible causes of inappropriate SPC and LPC selection include insufficient knowledge of drug characteristics and/or underestimation of potential risks. On the other hand, midline catheters exhibited the lower (30%) inappropriateness rate. Again, the presence of a VAT may play a role in explaining those differences. In fact, in the author’s hospital, SPC and LPC are placed by ward doctors and nurses, whereas midlines are placed by a VAT. As for CVADs, a VAT may be beneficial in guaranteeing a more correct VAD selection. Nevertheless, additional research is necessary to substantiate these findings. The main causes of PVAD inappropriateness may be related to prolonged dwell time beyond clinical needs, during which the device was either unused or no longer indicated; for instance, fourteen PVADs (4.7%) were used for chemotherapy, and six (2.0%) for total parenteral nutrition.

VAD failures varied by type. In fact, CVADs showed a higher failure rate than PVADs (26.1% vs 14.1%). This result, despite the lower appropriateness rate observed for CVADs, should be interpreted cautiously, given the small number of CVADs; further larger studies are needed to substantiate these findings.

Regarding PVADs, SPCs drove the PVAD failures mainly due to several episodes of malposition and phlebitis, which led to removal. Midlines registered no failures, supporting their role as a safe peripheral alternative when therapy exceeds short durations, even though this may also reflect selection bias and limited sample size. LPC also presented failures, even if only in 5 cases. SPCs and LPCs showed higher failure rates than midlines, aligning with their greater inappropriateness (59.5% and 43.8% vs 30% respectively).

Among CICCs a low event rates of complications were observed, being so consistent with their high appropriateness, and, likely, reflecting the standardized selection and placement by the VAT. Zero failures in CICCs is a reassuring result, but again, it is likely biased by the small number. The same for 100% failure among FICCs (3/3), which likely reflects poor insertion technique and management, but also may be biased by the very small sample.

Observing the distribution of VAD placement per DIVA and non-DIVA patients (Fig. 2), the trend suggests an increased number of VAD placements in the DIVA group, even if a statistical significance (p = 0.216) was not reached. Given the smaller sample size in this latter group (n = 25), these results should be considered with caution. Nonetheless, the relation between DIVA and the number of VAD places highlights a clinically relevant pattern that should be explored and statistically confirmed with larger studies.

As a matter of fact, inappropriate VAD selection is associated with higher complication rates, greater costs, and poorer patient experience (17, 18). Therefore, a proactive vascular planning leading to an optimal VAD selection should always be encouraged. As “antimicrobial stewardship” refers to a set of coordinated strategies aimed at improving the appropriate use of antibiotics (19), similarly, a “vascular access stewardship” should be implemented to select the right VAD for the right patient, at the right moment. Adequate training of healthcare providers and the implementation of standardized protocols for appropriate vascular access device selection represent crucial steps toward ensuring optimal clinical outcomes. Nevertheless, these measures alone may be insufficient to fully address issues of appropriateness and reduce complications in clinical practice. This underscores the need to explore the potential impact of specialized consultation from a dedicated Vascular Access Team (VAT). Investigating whether systematic VAT involvement can enhance decision-making processes, optimize device selection, and ultimately improve patient outcomes remains an essential area for further research. The current study will serve as a pilot study for a future controlled comparison between two hospital wards, one of which will receive regular input from a Vascular Access Team (VAT), and another operating without such structured support. Future multicentre trials are needed to validate these findings and evaluate the impact of dedicated vascular access teams on a broader scale.

This study has several limitations. First, it was conducted in a single centre, which may limit the generalizability of the findings. Additionally, the observational nature of the results does not allow for direct causality assessment regarding inappropriate VAD selection and complication rates. Lastly, the small number of CVADs does not allow a reliable comparison between the CVADs and PVADs’ appropriateness and to draw definite conclusions.

This study revealed a high prevalence of inappropriate vascular access device (VAD) selection among hospitalized patients, with PVADs being most frequently misused. Such inappropriate selection is likely to contribute to a higher incidence of complications, reduced clinical efficiency, and increased patient discomfort. These findings underscore the need for targeted interventions and institutional strategies to promote appropriate VAD selection based on clinical indication, therapy duration, and patient-specific factors. Further research is warranted to develop, implement, and evaluate evidence-based interventions that can optimize vascular access practices and improve patient outcomes.