Complications of peripheral intravenous catheters and risk factors for infiltration and phlebitis in children

Peripheral intravenous (IV) catheterisation is one of the most frequently carried out procedures in paediatric patients, and up to 50% of the hospitalised children are estimated to undergo this procedure (Ullman et al, 2020). These are not only critically ill children who require immediate IV antimicrobial treatment or rehydration therapy, such as patients suffering from sepsis, but also healthy children who need diagnostic tests.

Like many invasive procedures, the use of peripheral intravenous catheters (PIVCs) has a broad range of complications, including clotting, occlusion, leakage, infiltration, extravasation, phlebitis, partial dislodgement, accidental removal and infections (Lee et al, 2009; Rickard et al, 2012; Abolfotouh et al, 2014; Marsh et al, 2015; 2018; Webster et al, 2019).

Catheter-related complications are an important patient safety problem. Drug treatment may be postponed because of catheter-related complications, and catheters may need to be inserted repeatedly to continue treatment. These negative outcomes cause pain to patients and increase healthcare costs because of repeated procedures and prolonged hospital stays (Ullman et al, 2020). Approximately 30%–50% of PIVCs are removed early because of various complications (Helm et al, 2015).

In neonatal intensive care units, high complication rates of up to 97% of PIVCs have been reported (Legemaat et al, 2016; Marsh et al, 2021), while studies in older children reported lower rates (de Lima et al, 2011; Malyon et al, 2014). In a global study of paediatric peripheral IV catheter practice and performance conducted with 4206 children in 278 hospitals across 47 countries, PIVC complication risk factors were found to be: age >2 years; ambulance/emergency insertion; upper arm/antecubital placement; poor dressing integrity; and 24–72 hours dwell time (Ullman et al, 2020).

Although research has examined the incidence of infiltration, extravasation and phlebitis in children and neonates from different countries, to the authors' knowledge, no prospective study has focused on what the most common PIVC complications are in children in Turkey.

This 5-month prospective study aimed to identify the types of complications of PIVCs in hospitalised children and possible risk factors for the development of extravasation, infiltration and phlebitis.

Methods

Study design and settings

A descriptive, observational study was conducted over 5 months, from 1 July 2018 to 30 November 2018. It was carried out in the region's largest children's hospital, established in 1947, which has a bed capacity of 354, with 1400 employees providing care only to paediatric patients ranging from newborns to those aged 18 years.

The study included patients hospitalised in paediatric medical wards and cardiology, infectious disease and surgery wards. Any intensive care setting, including paediatric, neonatal, burn or cardiovascular units, was not included in the study.

During the study period, the departments included had a 160-bed capacity altogether and the total number of patients hospitalised in them per year was 2300.

Sampling

The incidence of catheter-related complications varies from one study to another. The infiltration/extravasation (I/E) rate was 5.5% in Özalp Gerçeker's et al (2018) study, and the phlebitis rate was 5% in Ben Abdelaziz's et al (2017) research. Using a calculation based on these rates, the sample size was determined to be 490 catheters to reach significance (type 1 error and 95% power)(Ben Abdelaziz et al, 2017; Özalp Gerçeker et al, 2018).

This study included 244 patients, 575 PIVCs and 1600 catheter days. The inclusion criteria were: being older than neonatal (4 weeks), being hospitalised and undergoing PIVC insertion from the first day of the study.

The exclusion criteria were: having undergone PIVC insertion before the study began, having a central venous catheter inserted, being transferred from an intensive care unit and having a PIVC inserted in the emergency department.

Data collection

After written consent had been obtained from the parents of children who had PIVCs, the patients were followed during their hospitalisation. Study participation was initiated by a member of the research team during routine morning visits.

Each patient had a case number, and each catheter was recorded on a form with no information that could have identified the patient.

The first author, who performed the observations, was a nurse trained in catheter-related complications who had worked in paediatric wards for 25 years. The visits were performed daily by the same researcher to prevent possible observer bias.

Every day, the researcher examined the catheter site for symptoms of PIVC complications and, if the PIVC was not in place, she noted the indication for removal. If infiltration/extravasation had occurred, the researcher also recorded the infiltrated fluid or drug and the flow rate.

The researchers were not involved in the patients' medical care or decisions to remove PIVCs or shift to central venous catheters.

Demographic and catheter characteristics were determined by the researchers using the ‘PIVC insertion site in paediatric patients evaluation’ form. If infiltration/extravasation developed, researchers noted the location with the ‘PIVC complications’ form. Both forms were developed by the researchers.

Data collection forms

PIVC insertion site in paediatric patients form

This form included the following possible risk factors (independent variables): sex; age; diagnosis; ward name and type; PIVC location (left or right, hand, arm and foot, or upper or lower extremity)(Lee et al, 2009; Malyon et al, 2014); time of PIVC insertion (by nursing shifts: 08.00-16.00; 16.00-23.59; 00.00–07.59); catheter gauge; presence of continuous IV fluid; and presence of IV drug therapy (Lee et al, 2009; Malyon et al, 2014; Gorski et al, 2016).

Continuous infusion was defined as the infusion of fluids to patients who needed to undergo continuous 24-hour fluid therapy for various reasons. IV drug therapy was defined as the administration of an antibiotic or other medication once or more times a day. These drugs were diluted and given over a short period (ranging from a few minutes to 30 minutes depending on the properties of the drug) by an IV fluid treatment pump.

No special technology was used to assist catheter insertion in any clinic participating in this study. All catheters were inserted by nurses working in those wards. The hospital where this study was conducted had no vascular access team.

PIVC complications form

This form, developed by the researchers, included questions about the location of I/E, PIVC insertion day, when I/E occurred, time I/E was noticed (time elapsed until the I/E), infiltrated or extravasated intravenous fluid(s) and medication, and interventions for I/E provided by nurses at the bedside.

The dependent variable was the occurrence of at least one PIVC complication, including extravasation, infiltration and phlebitis. Phlebitis was identified if pain or tenderness, erythema, warmth, swelling, induration, purulence or a palpable venous cord was present at the location of the PIVC. The researcher used the Phlebitis Scale and Visual Infusion Phlebitis Scale for grading phlebitis (Gorski et al, 2016).

Diagnosis of infiltration and extravasation was performed in accordance with the Infusion Nurses Society's (INS) Infusion Therapy Standards of Practice (INS, 2006). Among the symptoms of infiltration/extravasation are pain, oedema, fluid leakage from the puncture site, changes in colour including blanching from nonvesicant solutions (vesicants can produce redness; however, colour changes may not be visible if there is extravasation into deep tissue) and blister formation (Gorski et al, 2016).

The infiltration rate was calculated by dividing the number of infiltration incidents by the total number of peripheral catheter days in the study cohort. The infiltration percentage was calculated by dividing the number of infiltration incidents by the total number of peripheral catheters in the study cohort. The peripheral phlebitis percentage was calculated by dividing the number of phlebitis incidents by the total number of peripheral catheter days in the study cohort (INS, 2006; Gorski et al, 2016).

Statistics

The study data was analysed using IBM SPPS Statistics 17.0. The quantitative data were described using arithmetic mean and standard deviation, or median with interquartile range if they followed a non-normal distribution.

For categorical variables, percentages and frequencies were calculated, and the chi-squared test or Fisher's exact test was used for intergroup comparisons. The Holm-Bonferroni correction method was used to overcome familywise errors, and the Student's t-test or the non-parametric Mann-Whitney U test was used for the quantitative variables.

A multiple logistic regression test was calculated to predict the development of infiltration according to age, catheter size, whether the patient was hospitalised in the surgery or the paediatric ward, and whether or not the patient had a continuous infusion. The direct logistic regression was performed to assess the impact of several factors on the likelihood of phlebitis developing, and three independent variables (age in months, whether the patient was hospitalised in the surgery or the paediatric ward, and whether the catheter was placed in the antecubital fossa, hand or foot).

For the multivariate analysis, the authors included variables with a P value of <0.05 in the univariate analysis as well as any other variable predicted to be a potential risk factor for the outcome variables.

Research ethics

Before the study was conducted, written permission was obtained from the ethics committee and hospital management. Written consent was obtained from the mothers of the children whose catheters were monitored.

Results

Patient demographics and characteristics of peripheral venous catheters

The study included 244 patients, 575 PIVCs and 1600 peripheral venous catheter days. Of the 244 patients, 147 (60.2%) were male. The median age was 2 years (in a range of 1 month to 17 years). Nearly half of the patients (n=108; 44.3%) were aged <1 year, and 178 (66.0%) were aged <6 years.

One hundred and eighty-three patients (75.0%) were hospitalised on the paediatric medical ward, and 61 (25.0%) were on the paediatric surgery ward. The patients were admitted for conditions affecting the respiratory system (57 patients; 23.3%), cardiovascular system (50 patients; 20.4%) and gastrointestinal system (49 patients; 20.8%).

During the study period, 575 PIVCs were inserted. Of these, 292 (50.8%) were inserted on the right side of the body. The most used locations were veins in the upper extremities (n=410; 71.3%). The most used sites were the veins of the dorsum of the hand (n=240; 41.7%), the dorsum of the foot (n=103; 17.9%) the antecubital fossa (n=90; 15.7%) followed by other sites (Table 1).

Of the PIVCs, 71.0% (n=408) were 24 gauge in size. The length of the PIVCs was 19 mm (24 and 26 gauge) in 98.1% (n=564). Nearly 60.3% of the PIVCs (n=347) were inserted by nurses working on the daytime shift. Of the PIVCs, 83.0% were attached to infusion sets, 10.0% to serum sets and 7.0% to perfusion chamber sets.

Catheter mean survival time was 56.36±35.51 hours (ranging from 2 hours to 274 hours). During the study, 207 (36%) PIVCs were removed because of IV therapy completion, 305 (53.0%) owing to occlusion and 48 (8.3%) because they had become dislodged. The remaining indications for catheter removal were: patient transfer to the paediatric intensive care unit and shifting to central venous catheters (nine PIVCs;1.5%); phlebitis (one PIVC; 0.17%); the nurse observing problems with the line (two PIVCs; 0.34%); and patients or caregivers refusing treatment (three PIVCs; 0.51%).

Infiltration was observed at 8.7% (50) of the PIVCs, extravasation was observed at 0.2% (one) and phlebitis was observed at 15.8% (91). Most infiltrations were evaluated as grade I (84%) and phlebitis cases as grade I (54.9%)(Table 2) according to the INS infiltration scale (INS, 2006). The infiltration rate was calculated as 31.25 per 1000 catheter days.

Comparison of patients with and without infiltration

Infiltration developed in 50 of 575 of the PIVCs (8.6%). There were 25 (50.0%) boys in the infiltration group and 297 (56.6%) in the non-infiltration group; there was no significant difference between these two groups (P>0.05)(Table 3). The mean age of the patients with infiltration or without infiltration was 6.5 years and 4 years respectively, significantly higher in the infiltration group (P<0.05).

There was no significant difference in rates between the group that developed infiltration and the group that was free of infiltration by location (right or left; upper versus lower extremities; veins at the dorsum of the hand, foot and antecubital fossa (P>0.05 for each variable). The mean numbers of catheter days were 2.46±1.2 and 2.81±1.54 days in the infiltration group and non-infiltration group respectively, and there was no significant difference between the two groups (P>0.05).

The infiltration rate was significantly higher in the 22 and 24 gauge catheters (10.5%) than in the 26 gauge catheters (2.8.%)(P=0.0012). It was also significantly higher in the catheters with continuous infusions (12.7%) than in those in which continuous infusion was not applied (5.4%)(P<0.05).

The Hosmer-Lemeshow goodness of fit test indicated that the initial model, which included age, was a poor fit (P=0.016), so the age variable was eliminated from the final logistic regression analysis.

Direct logistic regression was performed to assess the impact of several factors on the likelihood of the catheters developing infiltration in the new model, which considered three independent variables (catheter size, whether the patient was in the surgery or paediatric ward, and whether the patient had continuous infusion or not). The full model containing all predictors was statistically significant (chi-squared: (3, n=575)=91.3; P<0.001).

Three of the independent variables made a statistically significant contribution to the outcome (catheter size, whether the patient was hospitalised at the surgery or paediatric ward, and whether the patient had continuous infusion or not) (Table 4). The strongest predictor of developing infiltration was catheter size. Study findings indicate that 22 and 24 gauge catheters would be 2.693 times more likely to develop infiltration than larger catheters. Being admitted to the surgical ward and having continuous infusion were also associated with nearly twice the risk of infiltration (Table 4)

Comparison of patients with and without phlebitis

During the study, phlebitis developed in 91 of the 575 PIVCs (15.8%). There were 59 (18.3%) male and 32 (12.6%) female patients in the phlebitis group, and there was no significant difference in terms of the variable of sex (P>0.05)(Table 5). The average age of the patients with PIVCs who developed phlebitis was 5 years 6 months and the average age of those who did not develop phlebitis was 4 years; age was significantly higher in the infiltration group (P<0.005).

There was no significant difference in the rates between the catheter group that developed phlebitis and the catheter group that was free of phlebitis in terms of catheter location (right or left; upper versus lower extremities; P>0.05 for each variable). The rate of phlebitis from catheters placed in the antecubital veins was significantly higher (25/90; 27.8%) than in the hands (38/320;11.9%) or feet (22/141; 15.6%)(P=0.002). The phlebitis rate was significantly higher in the surgery ward (24/100; 24%) than in the paediatric ward (67/475; 14.0%)(P=0.014). The mean numbers of catheter days were 3.09±1.8 and 2.73±1.45 in the phlebitis group and the non-infiltration group respectively, and there were no significant differences between the two groups (P>0.05).

The full model of logistic regression containing all predictors (age in months, whether the patient was hospitalised in the surgery or paediatric ward, and whether the catheter was placed in the antecubital fossa, hand or foot) was statistically significant (chi-squared: (5, n=433)=23.56, P<0.001).

As Table 4 shows, three of the independent variables made a statistically significant contribution to the model (age, administration to the surgery ward and placing the catheter in the antecubital vein).

Logistic regression revealed that PIVCs in the paediatric surgery wards were associated with a significantly higher risk of phlebitis than those in medical wards. In addition, logistic regression revealed that phlebitis was less likely to develop when PIVCs were inserted in the dorsum of the hand and foot rather than the antecubital fossa (Table 4).

Discussion

This 5-month prospective study investigated the complications and risk factors of PIVCs, including extravasation, infiltration and phlebitis in a tertiary paediatric centre. To the authors' knowledge, this study is one of the few long-term studies examining peripheral line complications in Turkey. During the study, infiltration was observed with 8.7% of the 575 PIVCs, extravasation was observed in 0.2% of patients, and phlebitis occurred in 15.8% of patients. The infiltration rate was calculated as 31.25 per 1000 catheter days.

In this study, infiltration and extravasation were observed with 8.7% and 0.2% of the PIVCs respectively, with the infiltration rate being 31.25 per 1000 catheter days. In one descriptive study from Turkey, which included 297 peripheral catheters, the infiltration rate was lower (2.9%) and the extravasation rate was higher (2.3%) than in our study (Temizsoy et al, 2017). The differences between the rates of infiltration and extravasation in these two studies might be because of their design and setting, differences between the institutions, health professional characteristics (such as the number of nurses per shift and the presence of an intravenous team).

In the present study, infiltration and extravasation rates were lower than in other studies (Taylor, 2015; Major and Huey, 2016; Park et al, 2016). Moreover, the results of these studies were measured as either percentages or infiltration numbers per patient or catheter days, which makes them hard to compare (Taylor, 2015; Major and Huey, 2016; Park et al, 2016; Temizsoy et al, 2017; Özalp Gerçeker et al, 2018). The infiltration rates were reported to vary between 0.2 and 25.2% (Taylor, 2015; Major and Huey, 2016; Park et al, 2016; Ben Abdelaziz et al, 2017; Özalp Gerçeker et al, 2018) while, according to a number of studies, lower rates were mostly achieved with quality improvement interventions (Tofani et al, 2012; Sriupayo et al, 2014; Duncan et al, 2018; Ray-Barruel et al, 2019).

In some studies, infiltration rates were reported to range from 5.5 to 59.7/1000 catheter days (Tofani et al, 2012; Miliani et al, 2017; Braga et al, 2018; Özalp Gerçeker et al, 2018). The variety in infiltration rates reported in different studies (Tripathi et al, 2008; Tofani et al, 2012; Miliani et al, 2017; Braga et al, 2018; Özalp Gerçeker et al, 2018) show that the infiltration and extravasation rates can vary from country to country, from state to state, from hospital to hospital and even between wards in the same hospital.

In this study, the main dependent predictors for the development of infiltration were catheter size (22 and 24 gauge versus 26 gauge), being in the surgical ward, and having continuous infusion via PIVC. The 22 and 24 gauge PIVCs were used mostly in older children, and had a higher flow rate (20-36 ml/minute) than 26 gauge PIVCs (13 ml/minute). De Lima Jacinto et al (2011) reported that the peripheral catheter size was not a risk factor for PIVC complications. However, a recent multicentre prospective study from Tunisia reported that the use of 24 gauge catheters was associated with PIVC complications (Ben Abdelaziz et al, 2017).

Although age was excluded from the multiple regression analysis because of poor fit, the mean age was significantly higher in the infiltration group, suggesting that it may play a role in the development of infiltration.

Being in the surgical department was an independent factor for the development of infiltration. Major and Huey (2016) reported that infiltration rates were higher in surgical than in medical paediatric wards (12.3 versus 9.6 per 1000 per catheter days), supporting the findings of the present study. Park et al (2016) reported that infiltration rates were higher in paediatric medical than in surgical wards (9.4 versus 2.8). These reports suggested that not only the age but also other associated risk factors might affect PIVC complications between hospitals.

In this study, phlebitis developed in 91 PIVCs (15.8%). Phlebitis rates have been reported in a range of 2–53% in different paediatric studies (Sriupayo et al, 2014; Monasor-Ortolá et al, 2019; Suliman et al, 2020).

In this study, the most important independent factors contributing to the development of phlebitis were hospitalisation in the surgery ward and age. One study found that catheter complications increased with age (Tripathi et al, 2008). This finding may be owing to the selection of the patients because 66.0% of patients were aged under 6 years, and the catheters movement depends on how well the fixation (securement) was performed.

Moreover, although no study has compared phlebitis rates in surgical and medical paediatric wards, the risk of developing phlebitis can be a marker for patient care and can vary between settings.

The present study found PIVC insertion at the antecubital fossa was a risk factor for developing phlebitis. This supports a previous study including adults, in which 815 PIVCs were reviewed (Miliani et al, 2017), although another multicentre study reported that PIVCs inserted in the antecubital fossa and forearm had a lower risk of phlebitis (Cicolini et al, 2014).

Limitations

There are some limitations to the present study.

First, it did not take into account some possible risk factors during insertion, such as the number of unsuccessful attempts before insertion, and the experience of the nurse.

Second, it was performed in a single children's hospital so there is a limitation to generalisability, reflected by the variety of the risk factors and the rate of PIVC complications.

However, to the authors' knowledge, this study is one of the limited number of prospective long-term studies focusing on PIVCs inserted in the children and gives additional useful data on PIVCs.

Conclusion

The results of this study demonstrated that the university hospital had relatively high rates of phlebitis and infiltration compared to those previously reported in the literature.

Older age and hospitalisation in the surgery department are common risk factors for developing infiltration and phlebitis.

The findings suggest that risk factors for PIVCs might vary, depending on individual inpatient wards, hospitals and countries. Awareness of potential risks is essential, because knowing about the present situation may lead to the development of care bundles that improve PIVC use. For the future, an international, user-friendly registry system for PIVC complications should be developed.

Because of the results of this research, another study was carried out to determine how the nurses working in the same hospital inserted PIVCs, what difficulties they experienced during insertion and how they approached complications. PIVC complications were considered as a patient safety problem; under plans drawn up by the hospital's quality management, it was decided to build a catheter team to standardise PIVC use. The nurses working at the hospital were trained in vascular access to achieve standardisation in their approach to catheterisation and complications.

A study is planned to determine the prevalence of PIVC catheter complications in different paediatric clinics and hospitals in Turkey. Quality improvement studies on PIVC insertion to prevent and reduce complications will continue.

KEY POINTS

• The use of peripheral intravenous catheters (PIVCs) has a range of complications, including clotting, occlusion, leakage, infiltration, extravasation, phlebitis, dislodgement, accidental removal and infection
• PIVC complications cause pain to patients and increase healthcare costs because of repeated dressing changes and prolonged hospital stays
• Catheter size, hospitalisation in the surgery ward and continuous infusion contributed to the development of infiltration
• Age, hospitalisation in the surgery ward and catheter placement in the antecubital vein contributed to the development for phlebitis.
• Being aware of PIVC complications and of what causes them should lead to better care bundles

CPD reflective questions

• What interventions do you implement in your clinic to prevent and reduce peripheral intravenous catheter (PIVC) complications?
• Which site for peripheral venous catheterisation do you prefer? Does your hospital have a written protocol for this?
• Do you monitor peripheral venous catheters in your hospital for complications such as phlebitis and infiltration?