Idle peripheral intravenous cannulation: an observational cohort study of pre-hospital and emergency department practices

Establishing peripheral intravenous catheter (PIVC) access is a fundamental element for facilitating modern day emergency health care [1]. It allows for the administration of a variety of symptom-relieving and/or potentially lifesaving infusates [2]. The ubiquitous use and perceived low risk associated with this device has led to frequent and routine insertion [3]. However, PIVC insertion is described as a traumatic, invasive procedure, with over half of patients finding the procedure at least moderately painful [4]. As 12–25% of first time attempts are unsuccessful [5, 6], the possible distress caused by failure of the first attempt is an important consideration in the clinical decision making whether a PIVC should be inserted.

Inflammatory complications are common, and studies have reported that 18–54% of PIVCs are associated with phlebitis [6‐10]. Serious infections can occur as the PIVC acts as a potential conduit for the introduction of infectious pathogens [11]. Previous research examining rates of intravascular related bloodstream infections (BSIs) found a pooled mean of 0.1% PIVC BSIs per 100 devices [12]. Staphylococcus aureus bacteraemia carries the highest morbidity and mortality rate of healthcare associated infections and is associated with up to 25% of all PIVC related bloodstream infections [13, 14]. The prolific use of the device implies many patients are suffering serious adverse events [15].

A fundamental concept when performing any procedure, including PIVC insertion, is to ensure potential benefits outweigh the risks. There are a wide range of clinical indications for PIVCs, making appropriate patient selection for insertion challenging. Although the clinical need for a PIVC is sometimes obvious, many are inserted in haemodynamically stable patients without a clear or immediate requirement. It could be hypothesised that junior clinicians may be tempted to insert a PIVC for the purpose of training or skills maintenance. Similar could be said for experienced clinicians who have seen the rapid and unexpected deterioration of patients. Clinicians may insert a PIVC ‘just in case,’ with negligible potential benefits, and studies report up to 50% remain idle and are never used [16, 17].

Peripheral intravenous catheter insertion in the ED has been associated with higher rates of idle PIVCs [17‐19]. The Cannulation Rates in the Emergency Department Intervention Trial (CREDIT) study conducted in Queensland, Australia highlighted that one-third of PIVCs inserted in the ED were idle [20]. Reducing rates of idle or never used PIVCs should be a goal of all emergency health care systems, including the pre-hospital setting. Paramedics are an important group of providers who are skilled in PIVC insertion and are commonly the first point of healthcare contact for the patient. Sparse knowledge of PIVC insertion practices in the pre-hospital environment exists [21].

The circumstances under which paramedics insert a PIVC and the number of attempts taken to complete the procedure have not been comprehensively explored within existing literature. Extrication difficulties, patient movement, bystanders and weather conditions are factors which may negatively impact successful PIVC insertion. The austere nature of pre-hospital care may lead to different decisions on when, or why, to insert a PIVC compared to clinicians in other settings such as the ED. This may have an effect on the proportion of idle PIVCs compared to ED insertion. The lack of literature regarding rates of idle PIVCs, especially in the pre-hospital setting has been highlighted previously [21].

We aimed to describe contemporaneous PIVC insertion practices in pre-hospital and ED patient cohorts, with a focus on idle PIVC rates. Specifically our objectives were to: (1) describe and compare the rate of idle PIVCs between the pre-hospital and ED setting; (2) compare pre-hospital and ED cannulation practices, and (3) identify patient and clinician factors associated with idle PIVCs.

The eligible patients and relevant exclusions are outlined in Fig. 1. Of the 1249 patients with PIVC insertions included (372 pre-hospital and 877 in ED), nearly half (45.2%) were over 60 years of age and half (50.1%) were female. Table 3 describes the characteristics of PIVCs placed in the ED, pre-hospital setting and combined, where Table 4 describes the pre-hospital setting only.

Across the emergency setting, including pre-hospital and ED, nearly 3 in 10 (29.3%) PIVCs placed remained idle. ED clinicians were 2.4 times more likely than paramedics to insert a PIVC that was not used within 24 hours of insertion. These findings suggest that paramedics in our study made sound decisions on when to insert a PIVC compared to hospital staff. Although this left 19.9% of pre-hospital PIVCs that were not used within the first 24 hours, our results for pre-hospital PIVC insertions align with recent literature that advocates for clinicians not to insert a PIVC unless they are over 80% certain it will be used [20]_.

We also demonstrated that a substantial proportion (39.5%) of PIVCs inserted by paramedics were unused while in the pre-hospital setting. Comparatively, we report a lower percentage of unused pre-hospital PIVCs than previous studies which report rates as high as 72–83% [24, 25]. The rate of unused pre-hospital PIVCs in this study is of importance and prompts questions surrounding paramedics’ motives for inserting pre-hospital PIVCs. Further investigation is warranted to achieve a comprehensive understanding of pre-hospital decision making regarding PIVC insertion.

The provision of symptomatic treatment in the pre-hospital environment is an essential aspect of patient care. Although analgesia provision is common and evidence-based [26], the role of antiemetics for treating nausea via the intravenous route is not as well defined [27]. We found that pre-hospital PIVCs were most frequently used for analgesia and treating nausea. Approximately half (49.8%) of all PIVCs inserted and used pre-hospital had intravenous anti-emetics (mainly ondansetron) administered through them; we therefore hypothesise that a number of these PIVCs were inserted solely for the purpose of administering anti-emetics. Since October 2017, paramedics in our setting have obtained access to ondansetron oral disintegrating tablets in addition to the long-standing availability of intramuscular anti-emetics. It is possible that a subset of PIVC insertions for administering anti-emetics were clinically not indicated, and even unnecessary.

For paramedics and ED staff in our study, the clinical guidelines for PIVC insertion are broad, and rely largely on clinician discretion. The number of PIVC insertion attempts differed between the pre-hospital and ED healthcare providers. Emergency department practitioners recorded more and a wider range of attempts than paramedics with ranges of 1–10 and 1–4 attempts respectively. Clinical guidelines for paramedics in our study recommend that a maximum of three PIVC insertion attempts are permitted per officer. It is likely that these guidelines, transport factors and time constraints consistent with the pre-hospital environment contributed to lower numbers of insertion attempts by paramedics. Conversely, many more attempts were taken for patients within the ED requiring intravenous access; this may be due to an emergent requirement for further diagnostic imaging or pharmacology.

Peripheral intravenous catheters inserted within the ED were frequently used for obtaining blood samples (96.1%) and in one-third of the patients, the PIVC was used for blood sampling only. This study excluded pathology collection as an appropriate PIVC use due to the procedure carrying increased risks to the patient and a higher rate of complications when compared with venepuncture [28]. It is likely that patients, especially with less urgent ATS categories within our sample had a PIVC inserted in the ED solely for the purpose of pathology collection. The large proportion of PIVCs being used for pathology collection in our study is congruent with other studies [16, 29]. Egerton-Warburton et al. reported a multimodal approach aimed at reducing inappropriate PIVC insertions [29]. This approach was successful in reducing unnecessary PIVCs, but not in lowering rates of PIVCs inserted for pathology collection only [29]. The financial burden and increased patient risks of PIVC placement as opposed to venepuncture must be carefully considered prior to deciding which procedure is to be performed. Previous research estimates that costs associated with PIVC insertion and removal range between A$13-$22; this does not include the costs related with the ongoing maintenance of the device [20, 30]. More importantly, patients with a PIVC are at continued risk of developing complications. These patients are also more likely to receive intravenous therapy that may not be appropriate [31]. It is predictable that these patients will have higher financial costs compared to those who receive venepuncture. Although pathology collection using a PIVC may be reasonable if intravenous medication, fluids or contrast is required, a PIVC for the sole use of specimen collection can lead to downstream adverse events [31]. Our study defined an idle PIVC as unused within 24 hours of insertion, excluding a normal saline flush or the collection of pathology via the PIVC. Previous studies have adopted various definitions of an idle PIVC; however, the most frequently used definition is consistent with ours, namely a PIVC that was inserted and remained unused [32]. It can be argued that clinically indicated PIVCs may become idle due to rapid changes in the patient’s clinical condition or haemodynamic status. The complexities of patient physiology may create uncertainty around the definition of an appropriate PIVC and a zero incidence of idleness is unrealistic. This imprecision has resulted in the absence of an accepted definition of a clinically indicated, yet unused PIVC. As such, we deemed any idle PIVC in our study as an inappropriate insertion. A recent systematic review highlighted the lack of literature surrounding the rate of idle PIVCs in the pre-hospital setting [21]. Our study is the first study to directly compare idle rates of PIVCs between pre-hospital and ED settings. This work demonstrates current practices of PIVC insertion across the broader emergency setting and lays foundations for future work about the clinical decision making of clinicians around PIVC insertion.

The rate of idle PIVCs within the broader emergency health care setting in this study was 29%. This study provides insight into pre-hospital and ED PIVC insertion practices and highlights the differences in rates of idle PIVCs between both cohorts, with PIVCs inserted in the ED 2.4 times more likely to remain idle compared to those inserted pre-hospital. Reasons for the differences in these practices are not well understood and requires more targeted research. We recommend a qualitative approach with clinicians of different backgrounds and experience to capture a more in-depth insight into clinician perceptions regarding PIVC insertion and use, guided by the findings of this study. Reducing the rate of idle PIVC insertion will lead to less risk and discomfort for patients and cost savings for healthcare systems; as such this should be the goal of all healthcare providers.