STRENGTHS AND LIMITATIONS OF THIS STUDY

The study was conducted in a high-complexity hospital.

This cross-sectional study included a large sample of patients treated in the emergency department (ED).

Patients were only followed up during their stay in the ED, which could have impacted the prevalence of peripheral venous catheter failure.

All data were collected retrospectively from electronic health records.

Catheter dwell time or drugs administered have not been considered in this study.

Introduction

The insertion of a peripheral venous catheter (PVC) is one of the most common invasive procedures performed by healthcare providers, being an indispensable tool in medical practice for the administration of fluids, blood derivatives, electrolytes or drugs, and other important clinical interventions.1 2 The use of a PVC may compromise patient safety, causing adverse effects such as phlebitis, occlusion or extravasation. There are currently high rates of these catheter-derived complications that affect millions of patients each year worldwide.3 Previous studies show that PVC failure is associated with length of hospital stay, mortality and the cost of hospitalisation.4 5

Non-standardised practices reported among healthcare professionals, among others, may play a significant role in the occurrence of these complications. One of the causes of this variability could be the poor training and the low levels of clinical guidelines implemented, that could have an impact in the prevention and the prior identification of PVC failure.6

The use of a PVC is a fundamental part of emergency healthcare. Over half of all emergency department (ED) patients will require a PVC during their visit.5 The evidence identified that the prevalence of unnecessary PVC ranges from 27% to 32% in ED.7–9 Additionally, in a recent systematic review, it was found that PVC failure due to infiltration and extravasation were higher in EDs compared with other healthcare settings (25.2% vs 12.3%).10 Although the success rates of the first insertion in the ED have been reported to be about 80%.11 12 Other studies have shown that ED was also a risk factor for other complications, such as suboptimal dressings13 or phlebitis.14 It is known that complex patients with multiple chronic conditions and psychosocial issues have more frequent interactions with healthcare systems, are more vulnerable to complications and are at higher risk for poor health outcomes.15 In addition, the number of older patients visiting the ED is increasing rapidly and, as a result, patients are becoming more fragile and complex.16

There is currently evidence on the association of PVC failure with factors such as catheter gauge, insertion site, catheter dwell time and drugs administered.14 17 18 Other studies have discussed patient-related factors such as age, sex or their chronic diseases.19–21 However, evidence regarding other sociocultural or emotional factors in the patient associated with PVC failure is scarce. In this context, the concept of care complexity plays a fundamental role by considering aspects related to the patient without limiting itself only to therapeutic aspects.22 In 2010, Juvé-Udina defined the care complexity individual factors (CCIFs) as a set of specific characteristics in each person related to the different determining axes. These may have complicate care delivery and contribute to adverse events.23

Previous studies carried out in public hospitals of Catalonia found that several CCIFs were associated with adverse events (pressure ulcers, falls and aspiration pneumonia),24 25 in-hospital mortality,25 hospital readmission26 and revisit to the ED at 30 days.27 In addition, recently a study also associated PVC failure with CCIF; however, this study was conducted in inpatient wards, without considering patients admitted in ED.28

Given the abundant use of PVCs in the ED, it is likely that many patients are suffering PVC-related complications. Therefore, the risk of PVC complications must be considered to improve patient outcomes. In order to help create strategies to reduce healthcare-acquired complications and strengthen safety culture, knowledge of the prevalence of PVC failure in the ED and its relationship with CCIF is essential.

MethodsObjective, study design and sample population

The aim of this study was to determine the prevalence of PVC failure and its association with CCIFs and other clinical and sociodemographic variables in patients admitted to the ED. This cross-sectional, descriptive-correlational study was performed to evaluate the association between PVC failure and CCIFs. This monocentric study was carried from 1 June 2021 to 30 June 2022. The setting of the study was the Bellvitge University Hospital, a tertiary care centre located in the southern metropolitan area of Barcelona, Spain. It is a referral high-tech hospital for more than 200 000 inhabitants. This facility has an ED equipped with 5 modules and 120 treatment rooms, and it attends approximately 118 000 patients annually. In addition, this department handles urgent cases of any specialty, except paediatrics and obstetrics.29

A consecutive sampling method was used. Systematic selection of all patients admitted to the ED during the inclusion period was conducted. The inclusion criteria of the study were patients with a complete nursing care plan in the electronic health records (EHR). Patients who did not have a PVC registered and all those under 16 years of age were excluded.

Data collection

Data were collected retrospectively from the EHR. The main outcomes were PVC failure and CCIFs. These data were recorded in the EHR using Architecture, Terminology, Interface, Information, Nursing and Knowledge (ATIC) terminology by the ED nurses responsible for each patient, and refer to the patient’s assessment, identified diagnoses and nursing interventions. For this study, we obtained this information from the clinical data warehouse of the Catalan Health Institute. Sociodemographic variables were collected from the Minimum Basic Data Set of the ED.30 The information from both sources was linked in this database through the patient episode numbers. All data were pseudonymised independently by the nursing knowledge management and information systems department using a unique identification number. In this way, the confidential data of the participants was preserved.

Measurements

The main variables of the study were PVC failure and 26 CCIFs. Both variables were identified and recorded in real time by the nurses in the nursing assessment form of EHR when the patient was attended to the ED.

PVC failure

The PVC failure was identified by nurses using ATIC terminology and were categorised into four main groups: phlebitis (phlebitis and signs of infection), extravasation (extravasation and infiltration), occlusion (by clots or other mechanical occlusion) and dysfunction (involuntary or accidental withdrawal, catheter dislodgement, dysfunction or leakage of fluid). Data regarding PVC failure were collected from the electronic nurse records, in which registered nurses reported any of the following nurse diagnoses during patient admission: catheter-associated phlebitis (code: 10001284); extravasation (code: 10002222); occlusion (code: 10005988) or dysfunction (code: 10005388); and dislodgement (code: 10017289) or involuntary withdrawal (code: 10010464/10010495/10010496), according to ATIC terminology.

Care complexity individual factors (CCIFs)

CCIFs are a group of patient characteristics related to different health dimensions that may complicate care delivery and contribute to poor health outcomes. CCIFs were identified by ED nurses and were classified into five domains: (1) comorbidity/complications, (2) developmental, (3) psycho-emotional, (4) mental-cognitive and (5) sociocultural. Patients were considered to have CCIF if they presented with at least one defined characteristic, as identified in a previous study.23 These CCIFs were collected from the nursing assessment e-charts as structured data based on the Architecture, Terminology, Interface, Knowledge terminology.31 The CCIFs included in this study are summarised in online supplemental file.

Clinical and sociodemographic variables

Haemodynamic instability was obtained from the early warning system named VIDA (the Catalan acronym for Surveillance and Identification of Acute Deterioration). This nursing surveillance improvement programme has evolved into an early warning score system that is used on a daily basis to assist clinical decision-making. The VIDA Score automatically classifies patients into five groups according to patient progress data: no risk (level 0), low risk (level 1), moderate risk (level 2), high risk (impending complication if not stabilised) (level 3) and manifested complication initial status (level 4). For the purposes of this study, the VIDA Score was classified as mild (levels 1–2) or high (levels 3–4) risk. Patients were classified according to the highest VIDA Score obtained during their visit. Patient progress data were extracted from anonymised EHRs and included: respiratory rate, oxygen saturation, temperature, mental status (level of awareness: 1=aware and orientated, >1=disturbed mental status), heart rate and systolic and diastolic blood pressure.24

Other clinical and sociodemographic variables such as age, sex, nursing care plan and triage level were also collected. The nursing care plan is an instrument for standardising clinical practice that includes nursing interventions and their programming, nursing diagnoses and clinical variables of the patients attended.32 On the other hand, triage level is the process that allows the patient to be assigned a level of prioritisation in clinical care. Nowadays, there are different validated triage scales, Spanish Triage System (SET) is the tool used in the hospital setting. SET System is based on the five levels of triage: 1—resuscitation, 2—emergency, 3—urgency, 4—less urgent and 5—non-urgent.33

Validity and reliability

CCIFs were collected according to the classification created in 2010 by Juvé-Udina et al. Through a participatory action research study involving more than 400 nurses from 8 public hospitals, CCIFs were identified and classified into a total of 5 domains. Each domain is structured into factors and specifications. These specifications were part of the coded and structured data in the initial and ongoing nursing assessment sections of the EHR, as described in the ATIC. This classification has been used in other studies to reveal associations between CCIFs and other unfavourable patient outcomes,27 and the predictive ability of this classification has already been demonstrated.24–26

ATIC terminology has been used since 2020 in ED records, although this terminology has been used in hospitalisation settings of Catalan Institute of Health since 2007, the major public healthcare provider in Catalonia (Spain). All nurses in the ED received training process, and they were provided on-site mentorship in clinical practice and methodology before implantation. In addition, superuser nurses are responsible for supporting the training process of all staff and provide support on the use of the information systems, in clinical analysis and discussions to improve nursing care provision.32

Patient and public involvement

None.

Statistical analysis

IBM SPSS V.25.0 statistical software was used. We reported descriptive statistics for clinical and demographic variables in the study. Qualitative variables (nominal, ordinal and dichotomous) were described using absolute frequencies and percentages; and quantitative variables were expressed as median and IQR. The χ2 test was used for the comparison of qualitative variables and the Student’s t-test or Mann-Whitney U test for continuous variables, depending on whether the data followed a normal distribution. The normality of data was evaluated using the Kolmogorov-Smirnov test. Inferential analysis was used to identify significant relationships between catheter failure and care complexity factors. The OR and CIs were calculated and a 95% CI was established for all cases. All statistical analyses were performed using two-tailed tests with an alpha error of 0.05, and a p value of less than 0.05 was considered significant.

Results

During the study period, 51 412 adult patients were admitted to an ED with a completed nursing health record. Of these, 35 968 (70%) patients with one or more PVC inserted during their ED visit were included. Table 1 shows the clinical and sociodemographic characteristics of the patients included in the study. The median age was 70 years (IQR 53.1–81.0) and 54.3% were male patients. The median of length of stay was 12 hours and triage level was ≤3 in more than 85% of cases (this level means patients require more urgent care). The main reasons for ED visits were: dyspnoea, malaise, COVID-19 infection, abdominal pain, chest pain and heart rhythm disorders.

Table 1

Patients’ demographic and clinical characteristics (n=35 968)

We observed that 80% (n=28 827) of patients had at least one CCIF. Table 2 presents the prevalence of CCIFs organised by domains of complexity. The most prevalent CCIFs were uncontrolled pain (32.8%), extreme age (39.5%), anxiety and fear (16%) and consciousness disorders (14.9%).

Table 2

Care complexity individual factors of adults in emergency department (n=35 968)

Among patients with a PVC, 0.9% (n=316) had some complication. The reasons for the PVC failure charted by the ED nurses were: dysfunction/occlusion 29% (n=92), extravasation/infiltration 26.3% (n=83), dislodgement/involuntary withdrawal 26.3% (n=83) and phlebitis 18.4% (n=58).

The association of CCIFs with PVC failure is summarised in table 3. The CCIFs associated with PVC failure were: incontinence (OR: 1.85; 95% CI: 1.39 to 2.45; p<0.05), haemodynamic instability (OR: 3.06; 95% CI: 2.41 to 3.88; p<0.05), transmissible infection (OR: 1.61; 95% CI: 1.11 to 2.32; p<0.05), vascular fragility (OR: 2.94; 95% CI: 2.09 to 4.15; p<0.05), fear and/or anxiety (OR: 1.45; 95% CI: 1.10 to 1.89; p<0.05), impaired adaptation (OR: 3.57; 95% CI: 2.17 to 5.86; p<0.05), consciousness disorders (OR: 2.76; 95% CI: 2.18 to 3.50; p<0.05), lack of caregiver support (OR: 4.28; 95% CI: 1.04 to 17.63; p<0.05) and agitation (OR: 4.07; 95% CI: 2.15 to 7.72; p<0.05). The median number of CCIFs was higher in patients with PVC failure than those without PVC failure (2 vs 1; p value=0.001). Moreover, table 4 shows the association of other clinical factors with PVC failure. The length of stay in the ED showed a statistically significant relationship with PVC failure.

Table 3

Association between peripheral venous catheter failure and care complexity individual factors

Table 4

Association between peripheral venous catheter failure and characteristics of emergency department patients

Finally, figure 1 shows that the prevalence of PVC failure showed an increasing trend in ED patients with a higher number of CCIFs (0.61% in patients with 0–1 CCIF, 0.89% in patients with 2–3 CCIFs, 1.78% in patients with 4–5 CCIFs, 2.58% in patients with 6–7 CCIFs and 6.5% in patients with >8 CCIFs).

Figure 1

Prevalence of peripheral venous catheter (PVC) failure according to the number of care complexity individual factor (CCIF).

Discussion

The prevalence of PVC failure in the ED was around 1%. The presence of a higher number of CCIFs was associated with PVC failure. Logistic regression identified the CCIFs significantly associated with PVC failure were incontinence, haemodynamic instability, transmissible infection, vascular fragility, anxiety and fear, impaired adaptation, consciousness disorders, lack of caregiver support and agitation. To date, there is the first study that evaluated CCIFs in the ED27; however, there have been other studies carried out in other departments (hospitalisation units).24–26 28 34 These previous inquires have found associations between CCIFs and other patient health outcomes such as in-hospital mortality, adverse events or hospital readmission. The results of the current study showed that 80% of the patients had CCIFs and that the mean number of CCIFs was approximately 2 per patient, consistent with previous studies.24–26 34

In relation to the number of patients with a PVC, the study showed that more than two-thirds of patients with a registered nursing care plan required a PVC, according to previous studies carried out in EDs.8 35 36 The prevalence of PVC failure was lower than other reports in this same department.10 19 This difference could be explained by different criteria from the data collection and the PVC complications. In this sense, in our study all data have been obtained from EHC, and possibly the number of PVC complications has been underestimated. Accordingly, previous studies in other settings reported variable rates,13 21 37 with values ranging from less than 1%38 to more than 50% of PVC failure.14 In addition, there is great variability when measuring and recording these complications. For example, in relation to phlebitis, at least 71 different phlebitis scales exist, with highly disparate criteria and minimal testing of validity.39 On the other hand, several studies have classified the complications of PVC in a similar way to the current study40–42; however, many of them only considered phlebitis and infectious complications, without considering other causes of PVC failure.20 43 44 In those studies where several PVC complications were considered, the highest rates were usually extravasation or phlebitis.5 10 21 45 In our study, the most prevalent cause was dysfunction/occlusion, followed by dislodgement and extravasation, and lastly, phlebitis. Phlebitis often takes hours or days to develop46; therefore, it is possible that the difference in the proportion of PVC failure in previous research is due to the patient follow-up period. Thus, the patient’s shorter ED dwell time may explain these results.

In relation to sociodemographic variables, we observed a relationship between PVC failure and age, consistent with previous studies.19 However, no statistically significant differences were found in relation to sex, although some studies have reported an association with female sex.21 28 47 The length of patient stay in the ED was related to PVC failure. Patients who spend the longest time in the ED are those who have higher PVC dwell time, these results are consistent with studies that have associated the catheter dwell time with the risk of complications.20 Our study found no association between the triage level and PVC failure, although another study found differences in risk of PVC failure and the different levels of triage.19 The triage level assigned in ED categorises the urgency of patients based on the reason for the consultation or severity, and determining their waiting time, but it does not reflect the care complexity. Therefore, within each triage level, there are heterogeneous patients with different CCIFs, which might explain our findings.

Regarding CCIFs and PVC failure, we observed the association with several care complexity factors. First, in the comorbidity/complications domain was associated the haemodynamic instability, incontinence, transmissible infection and vascular fragility factors. In relation to this last complexity factor, 4.5% of patients who visited the ED presented it. This indicates that a relatively high proportion of patients can have venous tortuosity or fragility, that could hinder the PVC insertion. Similarly, a recent study showed that the prevalence of patients with difficult access in ED was 8.9%.48 These results are in line with studies that have related the number of attempts to catheter insertion with the complications reported subsequently, thus confirming that difficult intravenous access is associated with more complications.14 37 Patients who are haemodynamically instability often require higher gauche catheters and higher levels of intravenous drug administration, which can contribute to PVC failure.14 17 Similarly, patients with transmissible infections required the administration of one or more intravenous antibiotics that can be associated with PVC failure.18 45 49 Other studies confirmed that age is an important factor to consider.19 Although the developmental domain was not associated with PVC failure, elders presented a higher frequency of PVC failure. The psycho-emotional domain was also associated with PVC failure, specifically the complexity factors of fear or anxiety and impaired adaptation. There is little evidence related to these factors and PVC failure. However, other studies show that fear and anxiety were associated with other unfavourable health outcomes, both in hospitalisation units and in EDs.24–27 In the mental-cognitive domain, consciousness disorders and agitation were found to be predictors of PVC failure. This relationship could be related to patient movement and catheter fixation and integrity. In recent years, it has been demonstrated that optimal fixations or reinforced dressings reduce complications, reducing movement. This causes less irritation of the endovascular tissue and reduces entry of bacteria into the wounded skin, among other benefits.13 50 Conversely, accidental or involuntary dislodgement has been one of the most prevalent causes of PVC failure.5 10 Finally, in relation to the sociocultural domain, the lack of caregiver support was related to the PVC failure, as shows in previous studies.28 Also, this CCIF was associated with other adverse events, such as pressure ulcer, falls or aspiration pneumonia.25 Specifically, in the ED, the figure of the caregiver or companion plays a fundamental role in improving communication between the professional and patient, as they actively participate in the care process and play a crucial role in decision-making.51 However, the benefits of caregivers in the ED remain underexplored and more evidence is needed.

In summary, the CCIFs associated with PVC failure were incontinence, haemodynamic instability, transmissible infection, vascular fragility, fear and/or anxiety, impaired adaptation, consciousness disorders, lack of caregiver support and agitation. Nevertheless, the evidence shows that in order to understand PVC failure must be considered other factors: (1) catheter factors, such as material, calibre or type of dressing used to fix it13 50; (2) clinical factors, such as days of catheter dwell, insertion site, length of stay or drug administration44 49; and also, (3) patient characteristics, such as age and vascular fragility.28

Limitations

This was an observational analysis with a large number of patients included in a referral hospital in Barcelona. In this study, we evaluated CCIFs that could be associated with PVC failure. Even so, there were some limitations to the current study. First of all, this study did not consider the complexity factor ‘major chronic disease’ because it was not possible to collect this data from the EHR. However, there is already evidence regarding the relationship between chronic diseases and catheter complications.20 Second, the prevalence of PVC failure may have been under-reported because the length of stay of patients in ED is relatively short. Patients are often transferred to inpatient units or other services, so a patient may have PVC failure that was recorded by a nurse in another department. We should take into account that EHRs in ED were only implemented a few years ago, which may have impact in the compliance of nursing records. In this sense, we relied on compliance in completing the EHR; however, since EHR are completed voluntary, some caution is required regarding interpretation. Finally, a cross-sectional design of our study limiting the ability to determine causal inferences, and we did not analyse other variables such as calibre catheter, catheter dwell time or drugs administered. Future research should be conducted using a longitudinal design to analyse the variables independently associated with PVC failure and thus provide more robust findings.

Conclusions

The CCIFs associated with PVC failure were: incontinence, haemodynamic instability, transmissible infection, vascular fragility, anxiety and fear, impaired adaptation, consciousness disorders, lack of caregiver support and agitation. Therefore, PVC failure was associated with several CCIFs-related comorbidity and complications, psycho-emotional and mental-cognitive domains. The prevalence of PVC failure increased as patients had a higher number of CCIFs. Therefore, complications derived from PVC are common and compromise patient safety. Knowing the causes associated with them could help to avoid unfavourable PVC-related health outcomes. Consequently, the early identification of PVC failure would help to stratify patients and implement preventive strategies.

Data availability statement

All data relevant to the study are included in the article or uploaded as online supplemental information. Not applicable.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

This study involves human participants. The study was carried out in accordance with current laws and regulations and was approved by the institutional review board of Bellvitge University Hospital (Ref. PR051/22). All data were pseudonymised independently by an individual outside the research team, who assigned a numerical code to each patient episode in an Excel database. In this way, the confidential data of the participants was preserved. All data were recruited retrospectively of electronic health records.

Acknowledgments

We thank the CERCA Programme/Generalitat de Catalunya for institutional support.