Ultrasound assessment of short peripheral catheter failure

Introduction

Short peripheral catheters (SPCs) are the most widely used venous access system. ¹ However, Marsh et al. ² noted that approximately one in three (36%) SPCs fail (defined as unplanned removal due to occlusion, infiltration, phlebitis, dislodgement, and infection). Thus, the high complication rate is the actual point of weakness of this device. Three pillars are well defined when considering the prevention of complications during SPC use: flushing with 0.9% Sodium Chloride solution before and after use (to access its correct functioning and to maintain its patency); ensuring aseptic technique during manipulation of the catheter and performing careful monitoring of the exit site to identify the early onset of inflammatory complications (such as phlebitis or thrombophlebitis). ³ Regarding the last point, the Visual Infusion Phlebitis score (VIP) is widely used as a scoring system (Supplemental File SFF1). ⁴ VIP was created by Jackson in 1998 ⁵ and includes six grades, from 0 (absence of phlebitis) to 5 (advanced thrombophlebitis), evaluating the presence or absence of specific signs: pain, erythema, swelling, induration, palpable venous cord, and pyrexia; a score >1 requires SPCs removal.

In the so-called “Ultrasound Era,” some clinicians have begun to explore the possibility of examining the exit site and the surrounding tissues with ultrasound, trying to identify the aspects that could predict the device’s failure. In 2021, Bahl et al. identified subcutaneous edema as an independent factor significantly correlated with SPC failure. ⁶ Interestingly, they reported that edema was noted about 29 h before the malfunction. In addition, the observations of Goel et al. ⁷ suggest that the appearance of “predictive” ultrasound signs (US) for failure (identified by them as luminal narrowing, increased wall thickening, and presence of thrombus) was not accompanied by observable signs during clinical examination.

Another complication easily observable with ultrasound is a fibroblastic sleeve (FS), a well-known complication reported in all mid and long-term catheters (e.g. Midline). ⁸ An FS reduces the ability to draw the blood sample and may be accompanied by infiltration. In some patients, arm edema may occur due to the obstruction of blood return to the heart. ⁹ These premises suggest that ultrasound could represent a valuable tool for monitoring exit sites and identifying early complications.

This prospective observational study aims to evaluate the sensitivity and specificity of the US and VIP in identifying early (pre-clinical) signs of peripheral venous catheter failure.

As a secondary outcome, we evaluate differences in time to positivity of US criteria for SPCs removal compared with VIP.

Material and methods

Ultrasound protocol

The US examinations were performed by a nursing student following a 20-h training program on ultrasound practice and the identification of phlebitis criteria defined as the study endpoint (described later). The Vascular Access Team of Sacco Hospital staff carried out the training program. On-site supervision by an expert ultrasound operator to confirm eventually undefined conditions was provided. The exams were conducted using the same ultrasound machine with a 4–12 MHz linear probe (Koninklijke Philips N.V., CX50 CompactXtreme, NL), cleaned, and disinfected following current recommendations before and after each use. ¹⁰

During the site scanning, the vein was examined in a short-axis view, and SPC images were taken in an out-of-plane view (Figure 1) using only B-mode images. The probe was utilized to compress the vein and assess thrombosis by its complete non-compressibility. The FS was diagnosed if the vein was partially compressible and a hyperechoic layer around the catheter was present without involving the vessel wall. ¹¹ Simultaneously, the subcutaneous tissue was assessed to exclude subcutaneous edema. The thickness between the posterior dermis layer and the anterior muscular fascia layer was measured, and the presence of echo-free spaces was observed to identify fluid presence (Figure 2).

OPEN IN VIEWER

Figure 1.

The vein was examined in a short-axis view and B-mode SPC images (out-of-plane view).

OPEN IN VIEWER

Figure 2.

Subcutaneous tissue evaluation (on the left). The thickness between the posterior dermis and anterior muscular fascia layers was measured, and echo-free spaces were used to detect fluid presence. Vein compressibility (on the right). The probe was utilized to compress the vein: thrombosis is detected by its complete non-compressibility. FS was diagnosed if the vein was partially compressible and a hyperechoic layer around the catheter was present without involving the vessel wall.

The US was judged positive if one or more of the following signs were present:

– subcutaneous edema (increase of thickness >2.5 mm compared to the first evaluation and/or presence of fluid in soft tissues);

– FS with a diameter >50% of the inner diameter of the vein;

– Thrombophlebitis (hyperechoic structure throughout the vascular lumen involving vessel wall and/or non-compressibility of the vein)

Results

A total of 200 patients who had SPCs inserted were enrolled. Four dropout patients were registered during the 4 days of clinical and ultrasound monitoring (one self-removing catheter at 24 h; one dislocated catheter at 48 h; two dislocated catheters at 72 h). During 697 evaluations, the nursing students sought ultrasound and vascular access expert assistance only 11 times. This accounts for just 1.6% of the total assessment. The student required the expert’s help to differentiate between FS and thrombosis. In all 11 consultations, the expert confirmed the student’s initial evaluation.

The median age was 70 (65–86), and 80 patients were males (40%). The study population and SPCs’ characteristics are reported in the Supplemental File SFF2.

At the end of the study, among 200 SPCs monitored, 94 (47%) did not experience any complication, 99 (49%) SPCs showed one or more of the three US signs (subcutaneous edema, FS, and/or thrombosis) leading to catheter failure within the first 120-h, while seven (3%) SPCs had echographic signs of thrombophlebitis. Still, they were not followed by catheter failure. Specifically, among these seven SPCs, six (83%) had FS, observed at 24 h for one case, at 72 h for three cases, and at 96 h for the remaining two, while one (17%) SPC had thrombosis observed by ultrasound at 48 h.

VIP assessment performed over 4 days showed that a score greater than zero had a significant correlation with SPC failure at 24 (p = 0.029), 48, 72, and 96 h (p < 0.001). Meanwhile, in the 4-day ultrasound examination, at least one ultrasound sign and FS alone significantly predicted potential complications at all checkpoints (p < 0.001 for both assessments over the 4 days). Additionally, the presence of ultrasound signs of thrombosis was significant at 48 and 72 h (p < 0.001 and p = 0.005, respectively), while edema was significantly correlated after 72 h (p = 0.018).

Positivity of the VIP, at least one of the three US (edema, FS, and thrombosis), and each single US at 24–48–72–96 h are reported in the Supplemental File SFF3.

Both VIP and US correlate with complications of SPCs at 24–48–72–96 h. However, in 28 cases (26.42% of complicated SPCs), VIP does not allow early recognition of phlebitis. Overall, VIP predicted the complication by 0.47 days (± 0.06 days), while almost one of the three US predicted the complications by 1.9 days (± 0.05 days).

The accuracy of both VIP and US was over 90% at 96 h. However, the ability to detect positive results gradually improved over time. VIP had a 6% sensibility on the first day, while US had 19%. By the second day, VIP had increased to 30% sensibility, and the US had improved to 72%. By the third and fourth days, VIP had reached 87% while the US had consistently achieved 97%. In all checkpoints, US accuracy is higher than the VIP described in Table 1.

OPEN IN VIEWER

Table 1.

Complications specificity, sensibility, disease prevalence, positive predictive value, negative predictive value, and accuracy; VIPS: Visual Infusion Phlebitis score >0; US: ultrasound signs (at least one of the three).

Timing and type of complication	Sensibility	Specificity	Disease prevalence	Positive predictive value	Negative predictive value	Accuracy
24 h VIPS	6% (2.23%–12.60%)	100% (96.38%–100%)	50%	100% (54.07%– 100.00%)	51.55% (50.31%–52.78%)	53% (45.83%–60.08%)
24 h Fibroblastic sleeve	16% (9.43%–24.68%)	100% (96.38%–100%)	50%	100% (79.41%–100.00%)	54.35% (52.22%–56.46%)	58% (50.83%–64.93%)
24 h US	19% (11.84%–28.07%)	100% (96.38%– 100.00%)	50%	100% (82.35%– 100.00%)	55.25% (52.89%–57.58%)	57.98% (52.35%–66.37%)
48 h VIPS	30% (21.24%–39.98%	100% (96.34%– 100.00%)	50%	100% (88.43%–100.00%)	58.82% (55.68%–61.89%)	65% (57.94%–71.61%)
48 h Fibroblastic sleeve	60% (49.72%–69.67%)	98.99% (94.50%–99.97%	50%	98.34% (68.98%–99.09%)	71.22% (66.04%–75.90%)	79.49% (73.21%–84.88%)
48 h Thrombosis	11% (5.62%–18.83%)	100% (96.34%– 100.00%)	50%	100% (71.51%–100.00%)	52.91% (51.19%–54.62%)	55.5% (48.31%–62.53%)
48 h US	72.00% (62.13%–80.52%)	98.99% (94.50%–99.97%)	50%	98.62% (90.99%–99.80%)	77.95% (72.07%–82.89%)	85.49% (79.82%–90.08%)
72 h VIPS	62.96% (51.51%–73.44%)	97.94% (92.75%–99.75%)	50%	96.83% (88.47%–99.18%)	72.56% (66.53%–77.87%	80.45% (73.85%–86.00%)
72 h Edema	6.17% (2.03%–13.82%)	100% (96.27%–100.00%)	50%	100% (47.82%–100.00%)	51.59% (50.20%–52.99%)	53.09% (45.48%–60.59%)
72 h Fibroblastic sleeve	97.33% (90.70%–99.68%)	95.88% (89.78%–98.87%)	50%	95.94% (90.03%–98.40%)	97.29% (90.15%–99.30%)	96.60% (92.69%–98.77%)
72 h Thrombosis	11.11% (5.21%–20.05%)	98.97% (94.39%–99.97%)	50%	91.51% (58.24%–98.81%)	52.68% (50.69%–54.66%	55.04% (47.42%–62.49%)
72 h US	100.00% (95.55%–100.00%)	94.85% (88.38%–98.31%)	50%	95.10% (89.20%–97.85%)	100.00% (96.07%–100.00%)	97.42% (93.89%–99.21%)
96 h VIPS	86.67% (69.28%–96.24%)	95.56% (89.01%–98.78%)	50%	95.12% (88.10%–98.09%)	87.76% (74.19%–94.70%)	91.11% (84.53%–95.54%
96 h Fibroblastic sleeve	93.33% (77.93%–99.18%)	93.33% (86.05%–97.51%)	50%	93.33% (86.53%–96.83%)	93.33% (78.57%–98.16%)	93.33% (87.29%–97.08%)
96 h US	96.67% (82.78%–99.92%)	92.22% (84.63%–96.82%)	50%	92.55% (85.88%–96.21%)	96.51% (80.09%–99.48%)	94.44% (88.71%–97.80%)

Discussion

Phlebitis is one of the most prevalent complications among patients with SPCs. However, due to the considerable variety of definitions and evaluation methods, more comprehensive ranges are reported in the literature.^{12 –14} Finding a standard and effective monitoring technique still challenges clinicians today.

In this study, we aimed to investigate the ability of ultrasound to identify early signs of phlebitis, defined as either the presence of a thrombus, FS, or subcutaneous edema. At all checkpoints, at least one of the three US signs correlates with SPC failure.

The US achieved an impressive accuracy rate of 94.44% at 96 h in predicting SPC failure. Moreover, in nearly one-third of complicated SPCs, VIP could not detect early signs of phlebitis. The US appears to be able to anticipate complications earlier than VIP, with a mean time of 1.9 days compared to 0.47 days, respectively.

Regarding US examination, it’s essential to consider signs other than thrombophlebitis, such as subcutaneous edema and FS.

It has been observed that subcutaneous edema is potentially correlated with catheter failure. The precise cause of edema in the subcutaneous tissue is still uncertain. As per Yabunaka et al., ¹⁵ it could be due to infusate triggered by infiltration in soft tissues. However, studies conducted on animals ¹⁶ and humans ¹⁷ have highlighted the possibility of it being associated with an inflammatory state, which may be caused by phlebitis. Although ultrasound may not be able to differentiate between the two conditions, our data emphasizes the significance of identifying subcutaneous edema to predict SPC failure.

Histological studies show that the FS is akin to a foreign body reaction, composed of a distinct connective tissue consisting of fibroblasts and collagen that envelops the catheter. ⁹ Theoretically, the FS surrounding a catheter can reduce the catheter-to-vein ratio and the blood flow, leading to catheter failure and thrombosis development. Recent research suggests this phenomenon has minimal to no effect on central venous catheters. ¹⁰ The results of this study indicate the opposite in the SPC setting, where the presence of FS is statistically correlated with device failure.

One exciting aspect of this study is that the protocol used was simple to learn and use. A short training was required to acquire the ultrasound examination skills. Our findings suggest that ultrasound can improve the insertion of SPC and monitor the exit site. The results of this study suggest that clinical evaluation and ultrasound are suitable options. Ultrasound is more effective, but operational difficulties may arise, such as the lack of equipment and staff training. For this reason, a risk-stratification approach could be considered, with particular attention given to patients with altered states of consciousness (e.g. sedated or unresponsive) and high-risk patients (e.g. pediatric and obese patients).

The study focused on SPCs, according to the recent taxonomy proposed by the Infusion Nurses Society. ⁴ The US could also be useful for evaluating other types of peripheral intravenous catheters, such as long peripheral catheters and midline. In this setting, clinical evaluation of the exit site is limited due to the catheter location in the deep vein of the arm, where only ultrasound can be used.

Limitations

One limitation of this study was its observational nature, which prevented clinical decisions regarding SPCs driven by US examination. As a result, it was impossible to determine whether early removal of the SPC before clinical signs led to resorption of the sheath, subcutaneous edema, and/or thrombus and subsequently in prompt healing of the involved vein. A potential area for future development could be to conduct a study that investigates this hypothesis.

Furthermore, no statistical inferences could be made regarding technical aspects of the insertion procedure, such as device choice and site of insertion. This is due to the decision not to correct for any confounding factors to increase the generalizability of the data obtained and preferring to focus on the evaluation of ultrasound efficacy.

Moreover, we set a 4-day period of observation that only allowed us to follow up with all SPCs for part of the duration of their service. However, the observed high specificity and predictive capacity suggest that the US could be influential throughout the indwelling time. Further prospective studies are needed to confirm this subject.

Conclusion

The presence of at least one between subcutaneous edema, FS, or thrombus is correlated to SPC failure. An ultrasound protocol that effectively detects these signs can evaluate SPCs and recognize device failure earlier than clinical observation. The learning curve for the ultrasound protocol is straightforward. This promising development dramatically enhances patient care, especially in cases where clinical monitoring is particularly challenging. These problematic cases may include high-risk patients, such as pediatric and obese patients, patients with altered states of consciousness, such as those who are sedated or unresponsive, and patients who require catheter insertion in deep veins. Future research will determine if ultrasound signs are a cause or result of phlebitis and SPC failure and if they correlate to an inflammatory state.

Supplemental Material