Article Text
PDF
Abstract
Objective Ultrasound (US)-guided peripheral IVs have a high failure rate. We explore the relationship between the quantity of catheter residing within the vein and the functionality of the catheter over time.
Methods This was a prospective, observational single-site study. Adult ED patients with US-guided IVs had the catheter visualised under ultrasound post-placement. IV placement time and catheter length residing in the vein was obtained. Exclusions included catheter not visualised, patient discharged from ED unless IV failed, <24 hour hospitalisation unless IV failed or patient self-removed IV.
Inpatient follow-up occurred within 24, 48 and 72 hours from the IV placement time. Catheter functionality was noted. If the catheter failed, the time and reason for failure was documented.
Results 113 patients were enrolled; 27 were excluded. Of the 86 study subjects, 29 (33.7%) patients’ IVs failed and 57 (66.3%) remained functional. Median time to IV failure was 15.6 hours. 100% of IVs failed when <30% of the catheter was in the vein; 32.4% of IVs failed when 30%–64% of the catheter was in the vein; no IVs failed when ≥65% of the catheter was in the vein (p<0.0002). The HR was 0.71 (95% CI 0.60 to 0.83), and for every 5% increase of catheter in vein, the hazard of the IV failing decreases by 29% (p<0.0001).
Conclusion The quantity of catheter residing in the vein is a key predictor of long-term functionality of US-guided IVs and is strongly associated with the hazard of failure within 72 hours. Catheter failure is high when <30% of the catheter resided in the vein. Optimum catheter survival occurs when ≥65% of the catheter is placed in the vein.
- ultrasound
Statistics from Altmetric.com
Key messages
What is already known on this subject
Ultrasound (US)-guided peripheral IVs are known to have a high failure rate after placement of 45%–56% compared with 19%–25% failure with traditional blind IVs. There is limited research on factors that should be modified to reduce ultrasound guided IV failure. We studied the relationship quantity of catheter that resides in the vein and catheter longevity.
What this study adds
The quantity of catheter residing in the vein is a key predictor of long-term functionality of US-guided IVs. Catheter failure is high when <30% of the catheter resides in the vein and optimum catheter survival occurs when ≥65% of the catheter is placed in the vein.
Introduction
Patients with poor intravenous (IV) access present a daily challenge to ED practitioners. Placement of an ultrasound (US)-guided peripheral IV catheter in this patient population is a viable and safe option. Successful cannulation with US-guided IV occurs in more than 90% of cases compared with 25%–35% with traditional IV placement in patients with difficult vascular access.1–3 Additionally, US-guided IV reduces the time needed for successful cannulation, improves patient satisfaction and requires fewer attempts.2 4 5
Once cannulated, the failure rate of IV catheters placed under US guidance is concerning compared with traditional blind IV placement. Overall failure rates after successful IV cannulation for US-guided IVs is 45%–56% when compared with traditional IV placement where failure occurs 19%–25% of the time.6–10 High IV failure rates impact patient care within the ED and negatively affects care once the patient has been hospitalised. The most common cause of US-guided IV failure is primarily due to infiltration, and other causes include catheter dislodgement and phlebitis.8 11 IV failure can result in a number of negative outcomes including interrupted medical therapy; painful phlebitis and reinsertions; increased hospital length of stay, morbidity and mortality from infections and/or skin necrosis from caustic medication infiltration; use of invasive procedures such as peripherally inserted central catheters (PICCs) and central lines and wasted medical/nursing time.12
In patients with difficult access, US-guided IVs are often the last recourse for IV access before resorting to more invasive procedures. Because failure rate is high, it is important to approach insertions methodically to improve survival rates. There are a few studied qualities of US-guided IV placement that have shown to increase catheter survival. Anatomic variations shown to increase US-guided IV longevity include placing catheters more distally in the arms and in veins that are shallow.11 Another study determined that using a catheter over a guidewire improves catheter longevity.9 However, these catheters are costly, may not be readily available at most EDs and may require sterile technique.
A practical variable that may alter the survival of US-guided IVs is the length of catheter that should reside in the vein. Currently, the general accepted rule is that an ‘adequate’ amount of the catheter should be in the vein to avoid failure of the catheter.13–16 However, this recommendation is only anecdotal and there has been no study to verify it. We propose a study to explore the relationship between US-guided IV catheter survival and the percent of catheter that resides in the vein.
Methods
Study design and setting
This study was a prospective, observational investigation of catheter longevity. It was conducted at a single-site, tertiary-care, Level I trauma centre with an annual ED census greater than 130 000 visits. This study was approved by the Institutional Review Board (IRB) at our institution. The IRB waived written consent for study participants and required that all subjects be verbally consented. The sample size was chosen based on enrolment feasibility and possibility of performing exploratory analyses; no prior sample size estimate was performed.
Ultrasound-guided IVs were placed by attending physicians, ED residents and a select group of trained ED ancillary staff (nurses and ED technicians) who were proficient in ultrasound-guided IV placement using single-user technique.
Selection of participants
The subject population consisted of a convenience sample of patients presenting to the ED. Initial screening took place by ED staff who then notified a researcher once a functional US-guided IV was placed in a patient with difficult access. Patients were consented by trained research staff and received an information sheet outlining the study protocol. Study participation was completely voluntary and consent was obtained prior to enrolment. Patient enrolment took place from November 2015 to June 2016.
Patients eligible for the study were required to be at least 18 years of age and have a functional US-guided IV placed in the ED. An IV was considered functional if staff was able to extract non-pulsatile blood and/or infuse a flush of normal saline without evidence of extravasation. Patients were excluded if (1) they did not meet inclusion criteria; (2) the IV was placed utilising a method other than US guidance; (3) the IV was not placed between the deltoid and wrist; (4) initial IV assessment could not be completed within 2 hours of IV placement; (5) the IV catheter could not be visualised during the initial assessment; (6) the IV catheter was not hubbed to the skin; (7) the patient was discharged from the ED, unless the IV failed; (8) the patient was hospitalised for less than 24 hours, unless the IV failed; (9) the patient self-removed his/her own catheter prior to discharge.
Initial assessment
The investigators carried out the initial assessment at bedside. A total of five physician investigators were trained on performing the bedside assessment and how to uniformly obtain measurements and capture images of the cannulated vein. Ultrasound assessment was performed using a high-frequency linear probe using one of the following machines: a Sonosite M-turbo, Phillips SparQ or Mindray Zonare. Still images were saved to Qpath, a Health Insurance Portability and Accountability Act of 1996 (HIPPA)compliant online point-of-care image archiving software. Ultrasound guidance was used to first visualise the cannulated vein in short axis and then identify the vein that was cannulated. The catheter was then visualised under US guidance in long axis, and the following measurements were obtained: the length of catheter that resided within the vein from the tip of the catheter to the point of insertion in the vein (see figure 1). Additionally, the diameter of the vein and depth of the vein at the point of catheter insertion was measured. These measurements were performed once by a single physician investigator.
(A) US image of cannulated vein prior to measurements (dotted white arrow pointing to catheter inside the vein, solid white arrows showing edges of vein). (B) US image of cannulated vein with measurements obtained; vein depth (solid white line) and length of catheter inside the vein (dotted white line).
The investigators also documented who placed the IV, the length and gauge of the IV catheter used, the time of IV placement, the time of IV assessment, the location on the arm where the IV was placed (upper arm, antecubital, lower arm), the identity of the vein cannulated, the presence or absence of an IV drip at the time of IV assessment, and the reason for the US-guided IV (history of poor IV access and/or multiple failed IV attempts during current visit). Data collected from the electronic medical record included age, gender, current Body Mass Index (BMI), most recent vitals at the time of IV assessment (BP, HR, RR and temperature) and medical history of end-stage renal disease, sickle cell disease, systemic lupus erythematous or IV drug abuse.
Follow-up assessment
Investigators performed a follow-up on the patient’s catheter within 24 hours, 48 hours and 72 hours from the time of IV placement. Two trained physician investigators performed all the follow-ups. At each follow-up time interval, the researcher noted the time of evaluation and assessed whether the catheter was still functional. A catheter was noted to be functional if the IV drew back 5 mL of blood, if the IV flushed with 5 mL of normal saline or if there was any IV fluids or medication actively dripping through the IV. Through chart review, the researcher also documented if a CT scan with IV contrast was performed prior to the time of follow-up assessment. If the catheter was identified to have failed during follow-up assessment, the date and time of failure and the reason for failure were documented. If the catheter failed or was removed prior to the follow-up assessment, then the IV failure time and reason was obtained through chart review or through discussions with the patient, family member or staff who had direct patient contact. If the patient was discharged prior to the time of follow-up assessment, then the time of discharge was documented and the IV was presumed functional until time of discharge unless otherwise noted in the chart.
Outcomes
Our primary outcome was to determine if a relationship exists between the percent of catheter length positioned in the vein relative to the overall catheter survival and failure. Secondary outcomes include assessing failed IVs and determining their time to failure and reason for failure and also determining if IVs injected with contrast dye from CT scan imaging affected IV survival.
Data analysis
We compared patients whose IV failed within the first 72 hours with those that did not. We examined categorical variables with Pearson’s χ2 where appropriate (expected frequency >5), otherwise Fisher’s exact tests were used. We examined age with a t-test, but the remaining continuous variables were examined with Wilcoxon rank-sum tests as the others were not quite normally distributed.
Two-sided p values <0.05 were considered statistically significant. A Cox proportional-hazards model was used to examine the time to failure. The SAS System for Windows V.9.3 (SAS Institute, Cary, North Carolina, USA) was used to calculate all statistical tests.
Results
A total of 113 patients were consented and enrolled in the study. Twenty-seven patients were excluded, leaving 86 patients in our study. Figure 2 illustrates the enrolment scheme with detail of the inclusion and exclusion groups. Only seven patients were excluded due to the inability to visualise the catheter. Twenty-nine (33.7%) patients’ IVs failed and 57 (66.3%) remained functional. Table 1 illustrates the patient demographics and shows there was no statistical difference in the patient population in the failed versus functional group with regard to age, gender, BMI, vital signs, reason for US-guided IV placement or patient comorbidities. Table 2 further shows there was no statistical difference in the two groups with regard to a drip being in place before/when the IV was assessed, the IV gauge, total catheter length and characteristics of the vein (location on arm, vein used, depth and diameter). Of all patients enrolled, 28 different providers placed IVs. Table 1 shows that there was no statistical difference when a nurse, ED technician, resident or attending physician placed an IV.
Flowchart demonstrating the distribution of patients.
Patient demographics
IV characteristics at time of initial assessment
Table 3 shows the statistically significant differences in the IVs that failed versus those that remained functional when evaluating the quantity of catheter dwelling in the vein. Less of the median catheter (44%) was dwelling in the vein in the failed group compared with the functional group, which had 53% of the catheter in the vein. When at least 50% of the catheter was placed within the vein, the majority (61%) of these IVs maintained patency. With a reduction in the quantity of catheter residing in the vein to less than 30%, there was a subsequent increase in IVs that failed at 100%. When the bulk of the catheter (≥65%) was placed in the vein, all the IVs stayed functional and none failed. Figure 3 demonstrates how the failure rate correlates with percentage of catheter in the vessel. Using a Cox regression model analysing the percent of catheter in vein, the HR was 0.71 (95% CI 0.60 to 0.83, p<0.0001); thereby for every 5% increase of catheter in vein, the hazard (or likelihood) of the IV failing decreased by 29%.
Trend of percent catheter in the vein and rate of failure.
Percent catheter in vein versus longevity
Of the 29 IVs that failed, the median time to failure was 15.6 (25th=3.7, 75th=30.8) hours. The median time to failure when the percent catheter in the vein was <30% and 30%–64% was 3.7 (1.7, 15.2) hours and 22.1 (8.7, 34.3) hours, respectively (p=0.09). Nine IVs (31%) failed in less than 5 hours, 11 (37.9%) failed between 5 and 24 hours, 7 (24.1%) failed between 24 and 48 hours, and 2 (6.9%) failed after 48 hours. The reasons for failure included infiltration,13 leaking,2 inability to flush,7 infiltration and leaking,1 inability to flush and infiltration,5 and leaking and inability to flush.1 No IVs failed due to infection or thrombosis.
A total of 15 patients received CT scans with IV contrast. Of the 57 patients whose IVs remained functional, 10 (17.5%) were injected with IV contrast and of the 29 patients whose IVs failed 5 (17.5%) received IV contrast (p=0.97).
Limitations
Our study was subject to a few limitations. It was performed at a single site and we were therefore unable to externally validate our findings. We are a tertiary referral centre with a highly complex patient population and frequently encounter difficult vascular access patients. Other sites may not have a similar frequency of vascular access issues, making our results potentially less generalisable.
There was no formal sample size calculation performed and enrolment was based on the feasibility of performing an exploratory analysis. Overall, the sample size was also small. Despite some compelling findings, it is possible that the study is underpowered. If so, the results may be exaggerated and there is a higher likelihood that the effect found may not represent a true effect.17
We only followed the functionality of the IVs up to 72 hours from the time of placement. Our study could have been more robust, had we followed these IVs for the entire duration of patient hospitalisations. The 72-hour timeline was selected based on the most current 2011 Centers for Disease Control and Prevention guidelines for the prevention of intravascular catheter-related infections, which recommended IV dwell times of 72–96 hours.18 At the time of study subject recruitment, our hospital IV policy also recommended removal of catheter within 72 hours. This policy has recently been modified. A recent Cochrane review recommended replacing IVs when clinically indicated rather than based on a strict time interval.19
Each catheter measurement was only performed once by a single physician investigator. We did not obtain multiple measurements of the same catheter from multiple investigators and therefore did not control for interobserver variability. However, to maintain consistency, all investigators underwent training to obtain measurements uniformly. In addition, the catheter is clearly visible and sonolucent to the observer, and any differences in measurements were unlikely to be substantial or change our conclusion.
Furthermore, we did not control for or gather data on the type of IV fluids, rate of fluids or type or absence of medications being infused as it relates to the longevity of the catheter. A multivariate study by Wallis et al showed that the following IV medication infusions increased the chance of catheter failure: IV antibiotics, IV hydrocortisone and IV antipyretics.20 Infusants have variable osmolarity and pH, which may impact catheter survival independent of percent of catheter in the vein.
Discussion
The inability to obtain IV access and any subsequent failure of these IVs presents a pressing problem in healthcare in that it is disruptive to the management of ED and hospitalised patients. This problem is further compounded in patients with poor access. Successfully placing IVs under ultrasound guidance solves the problem of obtaining access. However, once placed, US-guided IVs are known to have a poor shelf life and a high failure rate. Studies by Dargin et al and Elia et al show that US-guided IVs fail frequently 45% and 56% of the time, respectively.8 9 Furthermore, Dargin et al show that these IVs fail quickly in a median time of 26 hours, with 47% failing in the first 24 hours.8 Our data further support the poor long-term performance of US-guided IVs in which the IVs in our study had a comparable although lower failure rate of 33.7% and a median time to failure of 15.6 hours.
In order to combat the poor longevity of US-guided IVs, sonographers must make a concerted effort to focus on modifiable factors known to increase catheter survival. Fields et al identified two such factors that include placing IVs in arms veins located more distally in the antecubital fossa and forearm and also aiming for veins that are shallow with a depth <0.4 cm.11 Once an US-guided IV is in place, what other factors should be considered to ensure the longevity of the catheter? Our study is the first to our knowledge to take into account the percent of catheter dwelling in the vein as an indicator of catheter survival. Previously, the only available data were anecdotal and non-scientific references to an ‘adequate’ amount of catheter that should dwell in the vein.13–16
In our study, the highest failure was shown in those IVs in which less than 30% (or approximately one-third) of the catheter was placed within the vein and the highest survival occurred in those IVs in which more than 65% (or approximately two-thirds) of the catheter was in the vein. The clinical implications of this finding are multifold. Ultrasound-guided IVs that fail tended to do so quickly in which approximately a third of all IVs failed within the first 5 hours and those IVs with less than 30% (or approximately one-third) of the catheter in the vein failed in a median time of 3.7 hours. Patients who have IVs placed that have less than 30% (or approximately one-third) of the catheter dwelling in the vein should be considered for an alternative, more secure IV access such as a PICC line or a midline. Alternatively, a provider can consider placing a second US-guided IV once the patient is more volume repleted through IV fluids. Furthermore, providers must be mindful when signing these patients out to the admitting service because when an IV has been placed with less than 30% (or approximately one-third) of the catheter in the vein, it will be imperative to communicate that a new vascular access line will likely need to be placed.
Our recommendations are that sonographers make every attempt to place ≥65% (or approximately two-thirds) of the catheter within the vein so as to reduce the chances of failure and maintain catheter longevity. To accomplish this task, a calculation can be performed to determine the minimum total catheter length that should be used to cannulate a vein at a specific depth. Figure 4 illustrates this calculation, which is based on the Pythagorean theorem and assumes the following: a needle stick angle of 45 degrees and the distance the needle penetrates the skin from the transducer is equivalent to the depth of the vein. Given that the depth of the vein is a known variable, it is possible to determine the appropriate total catheter length required prior to starting the procedure to ensure that 65% of the catheter can be placed inside the vein. Adjustments to this model should be made if practitioners choose a different angle of insertion.
Illustration of the Pythagorean theorem (a2+b2=c2) used to calculate the minimum total catheter length needed to maintain 65% of the catheter inside of the vein. Based on a known vein depth (which represents ‘a2’), the assumption of a 45-degree needle stick angle and the assumption that the distance from the transducer of the needle penetrating the skin is equivalent to the vein depth (which represents ‘b2’).
Table 4 shows this calculation performed on a variety of depths and our recommended minimum total catheter lengths to be used for each depth. At many institutions, the longest standard angiocatheter that is available for US-guided IV placement is 1.88 inches, and based on table 4, when using this length catheter, the maximum vein depth that should be pursued for cannulation is 1.18 cm. For deeper vessels, longer catheters may be needed.
Calculated recommended minimum total catheter length based on a known vein depth (to maintain 65% of the catheter in the vein)
As part of this study, we also looked at whether IVs injected with IV contrast for CT scan imaging was a factor in IV failure. The purpose of this subanalysis was to determine if any specific uses of the IV had an effect on the longevity of the IV. Our sample size was small, but we showed no difference between IVs that survived versus those that failed when IV contrast was injected.
In summary, our study shows that the amount of catheter dwelling in the vein is an important factor that affects the failure and longevity of IV catheters placed under ultrasound guidance. The quantity of catheter residing in the vein is strongly associated with the hazard of failure within 72 hours. When an IV is placed where less than 30% (or approximately one-third) of the catheter is inside the vein, these IVs failed frequently and quickly. In order to reduce IV failure after IV placement, we recommend that more than 65% of the catheter should dwell within the vein. IVs injected with contrast for CAT scans did not alter the longevity of the catheter.
References
Footnotes
Contributors AVP and AB designed the trial. AVP, JT, ARB and AB supervised the conduct of the trial and data collection. AVP, JT, ARB and AB undertook recruitment of patients. AVP, JT and AB managed the data, including quality control. AVP and AB provided statistical advice on study design and analysed the data with assistance from the Research Institute. AVP, JT and AB drafted the manuscript, and all authors contributed substantially to its revision. AB takes responsibility for the paper as a whole.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent Obtained.
Ethics approval IRB at Beaumont Health System in Royal Oak, Michigan, USA.
Provenance and peer review Not commissioned; externally peer reviewed.
Presented at AVP, JT, ARB, B Hang, AB. Does the longevity of an intravenous catheter placed under ultrasound guidance correlate with the quantity of catheter that resides within the vein? American College of Emergency Physicians October 2016 National Meeting (Las Vegas, NV).
Linked Articles
- Primary survey