Utilization of a biomedical device (VeinViewer^®) to assist with peripheral intravenous catheter (PIV) insertion for pediatric nurses

Abstract

Purpose:

Vascular access in pediatric patients can be challenging even with the currently available technological resources. This nurse-driven research study explored time, cost, and resources for intravenous access to determine if a biomedical device, VeinViewer^® Vision, would facilitate improvements in pediatric access. In addition, this study looked at nurse perceptions of skills and confidence around intravenous insertion and if the use of the VeinViewer^® impacted these perceptions. Literature examining pediatric intravenous access success rates compared with nurse perceived skills and confidence is lacking.

Design:

Nonblinded randomized control trial of pediatric nurses working in an acute care hospital setting.

Methods:

A preliminary needs assessment solicited feedback from nurses regarding their practice, perceived skills, and confidence with placing peripheral intravenous catheters (PIVs). Due to the results of the preliminary needs assessment, a research study was designed and 40 nurses were recruited to participate. The nurses were randomized into either a VeinViewer^® or standard practice group. Nurse participants placed intravenous catheters on hospitalized pediatric patients using established procedures while tracking data for the study.

Results:

Needs assessment showed a majority of nurses felt a biomedical device would be helpful in building their intravenous insertion skills and their confidence. The study results did not demonstrate any clinically significant differences between VeinViewer^® use and standard practice for intravenous catheter insertion in pediatric patients for success of placement, number of attempts, or overall cost. In addition, no difference was noted between nurses in either group on perceived skills or confidence with insertion of PIVs.

Practice implications:

The ongoing need for resources focused on building nurse skills and confidence for PIV insertion was highlighted and organizations should continue to direct efforts toward developing skills and competency for staff that are responsible for pediatric vascular access. This study illustrates the importance of data-driven decision-making for expensive hospital-funded equipment purchases. This nursing led research study highlights how perceptions do not always align with outcomes. The lessons gleaned from this study may aid in decision-making around pediatric intravenous access practice.

1 ∣. INTRODUCTION

Vascular access presents unique challenges in the pediatric population because peripheral intravenous catheters (PIVs) are difficult to place in children (Crowley, 2003; Lininger, 2003). Intravenous access teams or specialists are sometimes used in hospitals and are a resource available to help start PIVs in the pediatric settings. Obtaining intravenous access is one of the most frequently performed procedures in hospitals; it is often perceived as routine and the impact on children can be overlooked (Kuensting et al., 2009). Furthermore, literature examining pediatric PIV success rates compared with nurse perceived skills/ability and confidence at placing PIVs on pediatric patients is lacking.

1.1 ∣. Review of literature

Additional studies have demonstrated the need for increased training and support for staff nurses to build upon their PIV placement skills as well as have highlighted the cost and success rates for insertion. A study of 656 PIV attempts at a pediatric hospital found a success rate of 44% for staff nurses, 23% for physicians, and 98% for the PIV nurse clinician. The average cost of placing a PIV in this study turned out to be $24.00, which led to an annual estimated cost for unsuccessful PIV attempts at that institution of $270,192 (Frey, 1998). Prior studies looked at 249 PIV attempts in a pediatric hospital and found that 53% were successful on the first attempt, 67% within two attempts, and 91% within four attempts (Lininger, 2003). Another study of 34 nurses who started 339 PIVs, found that the nurses age, years worked, number of PIV insertions per week, self-rated PIV skills, and specialty certification all resulted in a significant difference between successful and failed PIV insertions (Jacobson & Winslow, 2005).

There are a number of vein visualization devices on the market that can be used by the clinician to assist with PIV insertion. The VeinViewer^® is a noninvasive electronic visual aid device which utilizes visible and near-infrared light designed to illuminate and project an image of superficial subcutaneous vascular structures on the surface of the skin (Christie Medical Holdings, Inc., 2012). Initial research has shown that utilization of VeinViewer^® in a pediatric population to help insert PIVs significantly improved vascular access success by increasing the first-attempt success, reduced the number of total attempts per patient, and reduced the time required to successfully achieve access (Hess, 2010).

Other research has shown either no benefit or mixed results related to the use of VeinViewer^®. For instance, Perry, Chantal Caviness, and Hsu (2011) reported no difference in first attempt success rates between standard group and VeinViewer^® device group on patients 20 years of age and under. A total of 70% of nurses expressed neutral or unfavorable assessment of the device for children without chronic illness or dehydration, while 90% found the VeinViewer^® device to be a helpful tool for patients with veins that are difficult to access. Another study concluded there was no time difference in PIV placement, number of attempts, or pain scores between the VeinViewer^® device and the standard group in a tertiary pediatric emergency department on patients 0–17 years of age (Chapman, Sullivan, Pacheco, Draleau, & Becker, 2011). However, in subgroup analysis of children 0–2 years there was a decreased time to PIV placement with use of VeinViewer^® for PIV placement (Chapman, Sullivan, Pacheco, Draleau, & Becker, 2011). Another study was done in a pediatric hematology oncology clinic where 53 patients were randomized to the VeinViewer^® group or standard methods group for phlebotomy or intravenous access (Ramer et al., 2016). In this study, all nurses using the VeinViewer^® were experienced with venous access as a skill and the study showed decreased procedural time when using the VeinViewer^® versus standard practice. Subjects rated nurses as having more skill when using VeinViewer^® than those who did not use the device (Ramer et al., 2016).

In another study comparing three different vein visualization devices, the researchers found that suitable veins for cannulation were easily visible with the VeinViewer^®; however, the first attempt success rate was not significantly different between groups that utilized each of the different devices (De Graaff et al., 2013). Further, Peterson, Phillips, Truemper, and Agrawal (2012) found that unassisted methods proved to be more successful in PIV insertion success than the VeinViewer^® assisted methods. VeinViewer^® also decreased first attempt PIV insertion success among skilled nurses, 47% success versus 62% success by routine cannulation, in a study done by Szmuk, Steiner, Pop, Farrow-Gillespie, Mascha, and Sessler (2013). Vein visibility is enhanced, but near-infrared devices do not improve cannulation (De Graaff et al., 2013).

1.2 ∣. Theoretical framework

This study attempted to investigate a commonly occurring task, PIV insertion in hospitalized children, and determine how to improve the level of nurses’ perceived skill and confidence. The concepts included in this study are nursing skills and perceived confidence, PIV insertion, device utilization, nursing experience, and time and cost of PIV insertion. Two theories that fit well with this study are Bandura's (1977) theories involving self-efficacy looking at how one perceives their own ability to perform tasks and Benner's (1982) Novice to Expert theory evaluating the differences between newer nurses and nurses with more experience performing PIV placement skills.

2 ∣. PURPOSE

The current research site lacks a pediatric intravenous access team, instead there are one or two hospital-wide resource nurses that work each shift. The resource nurses are often viewed as having the highest proficiency in placing pediatric PIVs. A preliminary needs assessment found many staff nurses report that they are not confident or skilled in placing pediatric PIVs, and so they frequently rely on the resource nurses to assist with this skill. However, resource nurses have many responsibilities and may not be available in a timely fashion to assist staff nurses with PIV access, resulting in care delays. Since PIV access is a common and necessary procedure for many hospitalized pediatric patients, a small body of confident and skilled nurses may not be adequate to handle the high demand. In addition, limited resources result in a barrier to mentor staff nurses to develop adequate PIV skills.

The preliminary needs assessment, done prior to the start of this study, solicited feedback from nurses through a paper survey that consisted of 18 questions regarding nurses’ perceived skill/proficiency and confidence with placing PIVs, nurse demographics including years worked in nursing and years worked in pediatrics, clinical ladder level, frequency of placing PIVs, success rate of PIV starts, and if PIV access was problematic for themselves and their peers. Questions also addressed whether nurses believed a vein visualization biomedical device would be helpful. A total of 42% of the nurses on the inpatient medical unit at the study site completed the survey (N = 59). Ratings of PIV skill/proficiency and confidence with placing PIVs are shown in Tables 1 and 2. These preliminary results are described based on the respondent's level within the clinical ladder (professional career advancement program) with Clinical Nurse I (CN I) being the novice/new nurse to Clinical Nurse IV (CN IV) being the most experienced nurses. Results from the clinical coordinators (CC) and clinical resource nurses (CR) were combined with the most experienced nurses for data reporting. No respondents indicated they had “very high confidence” in starting PIVs. And 66% (N = 39) of the nurse respondents replied that “in the absence of support from other nurses” that PIV starts are a “moderate or serious problem” for themselves. A total of 79% (N = 47) responded that PIV starts are a “moderate” or “serious problem” for their peers “in the absence of support from other nurses”. And 72% (N = 41) of the nurses surveyed stated that they “believe a vein visualization biomedical device would be extremely helpful or very helpful” in building their PIV skills, while 73% (N = 43) felt it would help in “building their confidence” with placing PIVs. This preliminary information was used to determine the need for the initiation of the study further described here.

TABLE 1.

Respondents’ self-rating of PIV skill/proficiency by PCAP level

			PCAP Level
			Missing	CNI	CNII	CNIII/IV, CC/CR	Total
Self-rating of PIV skill/proficiency	ADVANCED Skill	Count	0	0	1	5	6
		% within PCAP Level	0.00%	0.00%	4.30%	20.00%	10.20%
	PROFICIENT Skill	Count	0	0	7	11	18
		% within PCAP Level	0.00%	0.00%	30.40%	44.00%	30.50%
	BASIC/NOVICE Skill	Count	1	8	14	8	31
		% within PCAP Level	100.00%	80.00%	60.90%	32.00%	52.50%
	NEVER started	Count	0	2	1	1	4
		% within PCAP Level	0.00%	20.00%	4.30%	4.00%	6.80%
	Total Count	1	10	23	25	59

TABLE 2.

Respondents’ self-rating of confidence in starting PIVs by PIDCAP level

			PCAP Level
			Missing	CNI	CNII	CNIII/IV, CC/CR	Total
Self-rating of confidence in starting PIVs	SUFFICIENT Confidence	Count	0	0	0	7	7
		% within PCAP Level	0.00%	0.00%	0.00%	28.00%	11.90%
	REASONABLE Confidence	Count	0	3	6	10	19
		% within PCAP Level	0.00%	30.00%	26.10%	40.00%	32.20%
	SOME Confidence	Count	1	2	7	5	15
		% within PCAP Level	100.00%	20.00%	30.40%	20.00%	25.40%
	VERY LOW Confidence	Count	0	5	9	3	17
		% within PCAP Level	0.00%	50.00%	39.10%	12.00%	28.80%
	Missing		0	0	1	0	1
	Total Count		1	10	23	25	59

The numbers of intravenous access assistive devices on the market is increasing, especially in hospital settings (Goren et al., 2001; Katsogridakis, Seshadri, Sullivan, & Walzman, 2008). The purpose of the current study was to determine if the VeinViewer^® device would be fiscally and clinically beneficial for use in an inpatient pediatric medical unit. The effect of the device on nurse skills and confidence in PIV catheter insertions was also investigated. Further, observations were made during data analysis based on years of experience since the study team was also interested in how nurses with varying levels of experience rated their skills and confidence and how use of a device may have impacted those perceptions.

2.1 ∣. Hypotheses

The hypotheses of the study include:

1. Utilization of VeinViewer^® will increase nurses’ perceived skills/ability in placing PIVs.

2. Utilization of VeinViewer^® will increase nurses’ perceived confidence in placing PIVs.

3. Utilization of VeinViewer^® will increase the success rate of initial PIV attempts compared to the standard practice.

4. Utilization of VeinViewer^® will decrease the number of overall PIV attempts per patient compared to the standard insertion practices.

5. Utilization of VeinViewer^® will decrease the overall cost of PIV insertions compared to the standard insertion practices.

6. Utilization of VeinViewer^® will decrease the overall time associated with PIV insertion compared to the standard practice.

3 ∣. METHODS

3.1 ∣. Design

This was a longitudinal experimental design with randomization of non-blinded subjects looking at PIV starts over time.

3.2 ∣. Subjects

Prior to implementation, the study was approved by the Colorado Multiple Institutional Review Board. Power analysis was completed to determine sample size using a method for time-series analysis developed by Hedeker, Gibbon, & Waternaux (1999). To detect a difference between the two groups, assuming two-tailed alpha of .05, power of .8, and a medium effect size, we needed a total of at least 200 insertions (100 in each group, five per participant). Registered nurses (RNs), working on the Inpatient Medical Unit were then recruited to participate in the study. Recruitment was done through flyers posted on the unit and via e-mails. The first 40 volunteers were selected for participation in the study. Prior to the start of the study, all participants were consented. Stratified random sampling, based on perceived skills and confidence at placing PIVs, was used to randomize the participants into the VeinViewer^® group or the standard practice group; 20 nurses were randomized into each group. All participants received general education on placing PIVs from Christie Medical staff. Then the nurses randomized to the VeinViewer^® group were trained on the use of the VeinViewer^® by a Christie Medical educator. The nurses assigned to the VeinViewer^® group were requested to utilize the VeinViewer^® for all PIV attempts during the study. The nurses assigned to standard practice were told to follow the organization's defined standard practice with all PIV starts during the study without the use of the VeinViewer^®.

3.3 ∣. Procedure

All participants were required to follow institutional policies and procedures for placement of PIV's. Data were collected through paper forms and surveys the subjects completed during the study. Data from these forms were then manually entered into a secure electronic database (REDCap) by the research team. One researcher would enter the data and a second member of the research team would validate the data and verify accuracy before marking the file as complete. All data were collected from the participants in this study over the course of 15 months. All participants worked on the medical inpatient unit.

3.4 ∣. Instruments

Prior to any data collection, nurses completed a prestudy survey to collect demographic data and to determine initial experience level with placing PIVs, perceived skill, and perceived confidence. This presurvey was developed by the research team. It consisted of four demographic questions and self-report of the participant's frequency of IV attempts and successful attempts, as well as one question on intravenous line (IV) skill/proficiency and one question on confidence in starting IVs. The demographic questions included the year they were first licensed as a nurse, how many years they worked as a nurse in pediatrics, how many years worked as a nurse on the current unit, and their level on the clinical ladder (Clinical Nurse I (CNI), Clinical Nurse II (CNII), Clinical Nurse III (CNIII), Clinical Nurse IV (CNIV)/CC/CR). The demographic form was also developed by the research team.

The participants were asked to complete two questions developed by the research team on perceived skills and confidence each month until the end of the study, at which time, all participants completed a final follow-up survey. The follow-up survey was given to participants upon completion, the timing of this varied by participant. Some participants finished in two months and the rest concluded when the study was stopped in the fifteenth month. The follow-up survey included all the same questions as the prestudy survey with four additional questions for only those nurses in the VeinViewer^® group. The extra questions were intended to obtain specific feedback on the nurses’ experiences with the VeinViewer^® device.

Information was collected by all participating nurses during each event and was tracked on a data collection tool (see Figure 1). This tool was created by the research team and collected data on time, number of attempts, staff resources, and supplies used for each event. An event was defined as the process for each nurse to attempt successful PIV insertion on one individual patient, to a maximum of two total attempts. Once a participant completed 10 events, they completed the study.

FIGURE 1.

Data collection tool

4 ∣. RESULTS

A total of 40 nurse participants were recruited for the study. Over the course of the study, seven participants dropped out prior to finishing the study, four due to relocation or transferring to another position, two did not attempt any PIV starts during the study, and one self-withdrew. Of the final 33 nurses remaining, six did not complete enough IV events to analyze so only 27 participant data sets were used for final statistical analysis. Cases were eliminated that completed less than half of all the IV events, which is less than 5 out of 10. Based on demographic analysis, exclusion of the seven participants increased the overall mean of number of years worked as an RN and years worked for the current unit in the entire sample, which makes us believe that the excluded nurses were less experienced then the retained nurses. Data describing the two groups based on years of experience and clinical ladder level (levels range from I to IV) are presented in Table 3. There were no meaningful differences in nursing experience between the two groups.

TABLE 3.

Demographics by group

Characteristic	VeinViewer® N (mean)	Standard practice N (mean)
#of years worked as RN in pediatrics	12 (4.17)	15 (3.98)
Years worked on current unit	12 (4.08)	15 (3.98)
Clinical ladder level	12 (2.67)	15 (2.73)

Patients with completed PIV attempts were very similar across nurse participant groups, with those patients in the standard practice group having a mean age of 8.87 years (SD = 6.79) and patients in the VeinViewer^® group with a mean age of 9.19 years (SD = 6.41). When it came to percentage of patients with difficult venous access in each group, again the two groups were very similar; 58% of patients in both groups were classified by the nurse as being a difficult stick.

All statistical analyses were performed using R statistical software. Comparisons were made between PIV starts completed utilizing the VeinViewer^® and those started utilizing standard practice using nonparametric statistical tests. Nonparametric tests were used due to small group sample sizes and deviations from normal distribution. Mann–Whitney U tests were used for all the continuous variables and Fisher's exact tests were used to test proportions (success rate).

The results of the statistical analysis are presented by examining each of the specific hypotheses of this study:

1. Utilization of the VeinViewer^® will increase nurses’ perceived skills/ability in placing PIVs.

   Figure 2 presents trend lines in the average perceived skill for both VeinViewer^® (red) and standard practice groups (blue). It appears that the average perceived skill decreased for both groups over the course of the study, but there were no significant differences in the average perceived skill between the two groups before (W = 78, p = .55) and after the study concluded (W = 83, p = .76).

2. Utilization of the VeinViewer^® will increase nurses’ perceived confidence in placing PIVs.

   Figure 3 presents trend lines in average perceived confidence for both VeinViewer^® (red) and standard practice groups (blue). Results suggested that perceived confidence fluctuated across the study with a steady decrease in average perceived confidence from time one through time 14, after which there was a brief increase in both groups through the end of the study. There were no significant differences in perceived confidence between the two groups before the study (W = 90, p = 1) and after the study concluded (W = 87, p = .56).

3. Utilization of the VeinViewer^® will increase the success rate of initial PIV attempts compared to standard practice.

   Figure 4 presents the trend in average success rate for both VeinViewer^® (red) and standard practice groups (blue). Although it appears that the standard practice group has an initially higher average success rates then the VeinViewer^® group, the difference was not statistically significant (p = .72). Further, although the average success rate in the VeinViewer^® group appears to be higher than in the standard practice group at the end of the study, this difference was not significant (p = .62).

4. Utilization of the VeinViewer^® will decrease the number of overall PIV attempts per patient compared to standard PIV insertion practices in absence of use of the VeinViewer^®.

   Figure 5 presents the trends in the average number of PIV attempts per patient for both VeinViewer^® (red) and standard practice groups (blue). As shown in the graph, the groups show very similar trends in the average number of PIV attempts across time, both before (p = 1) as well as after the study was completed (p = .62).

5. Utilization of the VeinViewer^® will decrease overall cost of PIV insertions compared to standard PIV insertion practices.

   Of the 2,286 items used for all study events as recorded on the data collection forms (Figure 1), the standard practice group used a total of 1,291 items (M = 86.07, SD = 19.84) and the VeinViewer^® group used a total of 995 items (M = 82.92, SD = 21.76). Further, the standard practice group patient total cost was $6,709.90 (M = 447.3, SD = 132.30) and hospital total cost was $1,370.43 (M = 91.36, SD = 13.51) across all events. The VeinViewer^® group total patient cost was $4,999.42 (M = 416.6, SD = 94.57) and total hospital cost was $1,098.50 (M = 91.54, SD = 18.95) across all events. There were neither differences between the groups in the average number of items used (W = 95, p = .83), nor the patient (W = 105, p = .49) or the hospital cost for the average number of items used (W = 90, P = 1).

6. Utilization of the VeinViewer^® will decrease the overall time associated with PIV insertion compared to standard PIV insertion practice.

   The standard practice group took a total of 545 min from the time of chlorhexidine scrub to the end time (M = 36.33, SD = 22.18) and the VeinViewer^® group took a total of 446 min (M = 37.17, SD = 19.81) across all completed events. The standard practice group took a total of 1,622 min to place the IV (M = 108.10, SD = 29.15) and the VeinViewer^® group took a total of 1,272 min (M = 106.40, SD = 38.68) across all completed events. There were no differences between the time of chlorhexidine to the end time (W = 83.5, p = .77) or the total time to place the IV (W = 99, p = .68).

FIGURE 2.

Perceived skills/ability across study period

FIGURE 3.

Perceived confidence across study period

FIGURE 4.

IV insertion success rate

FIGURE 5.

Number of IV attempts per patient

An additional finding, nurses lower on the clinical ladder had higher self-perceived skill and confidence than those more advanced on the clinical ladder. Also, nurses utilizing the VeinViewer^® initially had lower success rates than those in the standard practice group.

5 ∣. DISCUSSION

The VeinViewer^® group did not differ from the standard practice group on any of the indicators (perceived skill, perceived confidence, number of attempts, IV success, time to insertion, and cost). Nurses in the VeinViewer^® group had initial lower success rates than those in the standard practice group which was possibly indicative of a learning curve for those using the device. There also were no significant savings in insertion time or cost as a result of using VeinViewer^®.

Nurse participation in the current study was voluntary, which increased chances of self-selection bias. It is possible that some nurses with little experience or low confidence at starting PIVs may have chosen to avoid participation in the study. Also, those nurses that volunteered for the study may not have been representative of the overall nursing population on this unit. Even though the study was extended, in order to collect as many responses as possible, attrition was another unforeseen limitation in our study. This led to our final sample size being smaller than anticipated. Additionally, this study was only completed with nurses at one pediatric institution and on a general inpatient medical floor; therefore, results cannot be generalized beyond this population. Finally, the research team only evaluated standard practice participants against one biomedical device and did not compare other devices.

Upon completion of the study and review of the literature, the team determined further research is needed to provide additional evidence on the benefits and drawbacks (pros and cons) associated with the use of vein visualization devices, including VeinViewer^®. Research on the cost associated with the use of these devices in hospital settings is needed in order to better understand the implications of incorporating these devices into practice. Additionally, there may be some benefit of studying more specific patient subsets to determine usefulness of this device (i.e., patients with history of difficult PIV access, infants and toddlers, and patients with chronic illness).

6 ∣. CONCLUSION

This study compared research groups across time and on a variety of indicators which allowed the research team to be able to more completely analyze the data. Initial interest in use of the device on the pediatric medical unit was generated by staff nurses. Through a several year process the study results showed that based on cost of the device and no clinically significant differences between VeinViewer^® aided PIV access and standard practice there was not enough evidence to support the purchase of the device for use in this setting.

The knowledge gained from this study provided valuable understanding into the current status of nurses’ skills and confidence with PIV insertions on this unit as well as general success rates and use of other nursing resources for this skill. It is clear that a small subset of our nursing staff are really comfortable with this task and that often times nurses seek out other resources to assist with PIV insertion. Our organization has been looking into the possibility of PIV access teams as well as other vein visualization devices since this study has completed.

The ongoing need for resources focused on building nurse skills and confidence for PIV insertion was highlighted in this study and organizations should continue to direct efforts towards developing skills and competency for staff that are responsible for pediatric vascular access. Finally, this study illustrates the importance of data-driven decision making for expensive hospital funded equipment purchases.

6.1 ∣. How might this information affect nursing practice?

The information from this study can help nurses decide if they should utilize assistive devices when starting peripheral intravenous lines. The research done at this institution helps others to consider implications for the use of assistive devices in nurses with varying levels of IV placement experience. Additionally, this research can help shed light on some cost implications of purchasing an assistive device and the time resources needed and costs incurred during placement of intravenous lines. As technology continues to advance and new devices are offered for use in healthcare settings, it is important that nurses consider the evidence that is available for those devices and think broadly about their use, cost, and practice implications before implementation.