Access time and complications in central venous catheter insertion in patients presenting to the emergency department of a tertiary hospital: A prospective cohort

Abstract

Objective: This study compares the access time and procedural-related complications in CVC insertion via ultrasound guidance versus the landmark technique. Methods: A prospective cohort study was performed in the emergency department of a tertiary care hospital from January 2016 to December 2018. Patients with the indication of CVC insertion were included. Primary outcomes included access time and total procedure time; secondary outcomes included post-procedural complications. Seldinger’s Technique was adopted for CVC insertion in both groups. For qualitative variables, chi-squared and Fisher’s exact tests were applied, and quantitative variables were reported as mean and standard deviation. Differences were assessed by the independent t-test and the Mann-Whitney test. Results: A total of 252 patients were enrolled and divided into exposed (ultrasound-guided technique) and unexposed (landmark guided technique) groups. Insertion of CVC on the first attempt was successful in 82 patients (62.1%) in the ultrasound group and in 68 patients (56.7%) in the landmark group. The mean access time (seconds) in the exposed group was 21.9 and in the unexposed group was 29.9 with a mean difference of only 8 seconds. The total procedural time (in minutes) was 32.2 versus 22.7 sequentially with a mean difference of 9.5 minutes. Fewer venipuncture attempts and arterial pricks were recorded in the exposed group (1.4 and 10/132) as compared to the unexposed group (1.8 and 36/120). Conclusion: Ultrasonographical-inserted CVC had a higher success rate with fewer post-procedural complications but required longer total procedural time with inappreciable access time.

Keywords:

• central venous catheter
• ultrasound
• landmark technique
• emergency
• internal jugular vein

1. Introduction

The prime objective for a positive outcome of a critically ill patient in the Emergency Department (ED) remains time-sensitive management (Mittal & Coopersmith, 2014). Intravenous access lines are pertinent to patient care due to the necessity to deliver parenteral medications, nutrition, fluids, and blood products (Panepinto et al., 2021). Around 60% to 90% of patients will need a peripheral intravenous line (PIV) during treatment (Panepinto et al., 2021). Central venous catheter (CVC) insertion in the ED is often inevitable due to the urgent need for resuscitation, invasive hemodynamic monitoring, and the requirement of inotropic support or placement of a pacemaker. It is sometimes the only venous access available (Miller, 2002; Saul et al., 2015). Although CVC insertion can be a chief component in the treatment of a critically ill patient, it is also associated with mechanical (5–19%), infectious (5–26%), or thromboembolic (2–26%) complications that may affect morbidity and mortality (McGee & Gould, 2003; Merrer et al., 2001; Richards et al., 1999). It is important to evaluate the safest technique for the placement of CVC without compromising the quality of health care.

Commonly used sites for CVC are the internal jugular, subclavian, and femoral. The former two are preferred in the ED due to the advantage of invasive hemodynamic monitoring and less risk of local infection and thromboembolic phenomena (Merrer et al., 2001; Steele & Irvin, 2001). Although internal jugular vein catheterization is preferred, its complications are well-documented and cannot be disregarded. Complications may include internal carotid artery puncture, hematoma formation, pneumothorax, or hemothorax (Miller, 2002).

There are two techniques of CVC insertion: landmark-guided CVC insertion (LCT) and ultrasound-guided CVC insertion (UCT). Generally, LCT placement is utilized with the reliance on the expertise of the anatomic structures and the palpation of the arteries. Unfortunately, this technique cannot consider all the anatomic differences at the insertion point (Saugel et al., 2017). The use of UCT has exponentially increased in the hospital setting due to its advantages, including low cost, ease of use, and portability (Hoffman et al., 2017). Visualizing the intended central vein and its surrounding anatomical structures utilizing an ultrasound can improve placement accuracy, reduce the risk of complications, and create a better patient outcome (Hoffman et al., 2017). Complication rates vary depending on operator experience and the patient’s co-morbidities (i.e., coagulopathy, hemodynamic instability). In addition, CVC may fail if the vein is thrombosed or if anomalies in the patient’s anatomy cause the operator to pass the needle in an inappropriate direction. The current literature comparing the two techniques has demonstrated that using UCT decreases the number of attempts and total procedure time (Froehlich et al., 2009). Jijeh et al. conducted a study to compare LCT to UCT. The UCT placement on patients had a 100% success rate, whereas the LCT placement only had a success rate of 77%. Patients receiving the UCT also had a better outcome due to the absence of punctures in the carotid artery compared to the LCT group, which had a carotid artery puncture rate of 25% (Hoffman et al., 2017). Leung et al. also demonstrated that UCT had a lower complication rate and higher successful insertion rate than LCT. Many meta-analyses have also concluded that UCT is a more successful technique as compared to LCT using the data from high income countries (HICs). Very few studies have mentioned the comparison of UCT versus LCT in the ED setting. There is no evidence of conducting such study in Emergency Department of Pakistan. The purpose of this study is to compare the access time and the total procedural time of CVC administration by the techniques and to assess the post-procedural complications caused by each in patients presenting to the ED.

2. Methods

2.1. Study design

We conducted a prospective cohort study to compare the access time and procedural-related complications in CVC insertion via ultrasound guidance versus the landmark technique from January 2016 to December 2018.

3. Study settings

This study was conducted at the Aga Khan University Hospital (AKUH) ED in Karachi, Pakistan. Recruitment employed a purposive sampling technique of patients visiting the AKUH ED who fulfilled the study’s eligibility criteria. Patients were divided into two groups: exposed patients received the UCT procedure, and unexposed patients received the LCT procedure. We required a minimum sample size of 560 patients, with 280 in each group, to achieve 80% power with an anticipated difference in mean access time among the two groups of negative 19 seconds (Filho LP et al., 2013) and a significance level of 5%.

Patients aged 16 years and above requiring CVC insertion during their management in the ED were included in this study. We excluded patients with a query of cervical spine injury, blunt or penetrating chest trauma, chest drain due to any underlying lung pathology, deranged coagulation profile that could not be corrected with a blood product transfusion such as Fresh Frozen Plasma (FFPs), or platelets, or risk of increased bleeding due to underlying pathology (Leung et al., 2006).

For LCT, access time was recorded from the time the finder needle was introduced to locate the internal jugular vein (IJV) until the aspiration of venous blood in the large introducer needle was successfully attained. For UCT, the time started when the ultra-sonographic probe was placed on the neck until the venous blood was aspirated in the introducer needle.

Emergency physicians, residents, or medical officers working in the ED who had successfully performed more than 25 CVCs without supervision implemented the procedure for evaluation of this study. The procedure was considered successful when the aspiration of venous blood was achieved in the introducer needle. The procedure was labeled unsuccessful if the initial method was unsuccessful after three attempts. An attempt was defined as the introducer needle’s entry into the skin and removal from the skin. After three failed attempts of the landmark technique, the fourth attempt was made via ultrasound-guided technique to avoid complications. If both techniques failed, then an alternative site was obtained and documented. Alternative access sites included the contralateral IJV, the subclavian vein, or the femoral vein (Leung et al., 2006).

Similarly, if one physician failed to complete both techniques, another assisted. Access time and total procedure time were recorded for both techniques. Total procedure time commenced from the preparation of the patient for CVC insertion, which included positioning, identification of the landmarks, arrangement of instruments and ultrasound machine, and successful insertion of the CVC until the dressing application following the procedure. The beginning of access time was counted when the probe was placed on the patient’s neck to locate the IJV; however, the time utilized for the ultrasound machine’s setup and the probe’s preparation was not included. Furthermore, end time was defined as the minute aspiration of venous blood was achieved.

After the procedure, a trained operator recorded the data, including the patient’s comorbidities, indication for CVC insertion, and the method used. Assigned nursing staff recorded the primary outcome (access time and total procedure time). Secondary outcomes (post-procedure complications, like pneumothorax, hemothorax, and malposition) were assessed by plain chest radiograph, a routine investigation after CVC insertion that was reviewed by an on-call consultant and counter-checked with the radiologist’s final report. Arterial puncture, localized hematoma, and multiple attempts were assessed clinically.

Frequencies and percentages were reported for qualitative variables (such as gender and complications) and were assessed by chi-squared and Fisher’s exact tests, where appropriate. Quantitative variables (such as age, access time, and total procedure time) were reported as mean and standard deviation. Shapiro-wilk test was applied to check the normality of the quantitative data. And differences were assessed by independent t-test and Mann-Whitney test. A p-value of<0.05 was considered significant throughout the study. Stata (version 15.0) was used for data analysis.

4. Results

A total of 252 patients participated in this study, with 132 in the exposed group and 120 in the unexposed group. The exposed group had marginally more females (60.8%, n = 73) than males (51.5%, n = 68), with an overall mean age of 52.64 ± 12.3 years and a BMI of 24.7 ± 5.3. The unexposed group consisted of more males (60%, n = 73) than females (39.1%, n = 47), with a mean age of 53.26 ± 12.3 years and a BMI of 23.1 ± 5.9. No statistically significant difference was noted in the demographics of the two groups, but it was observed that a higher proportion of males underwent LCT (60.8%) while more females experienced UCT (51.5%).

Most patients were hypotensive (p = 1.01) in both groups; therefore, the prime objective of catheterization was delivery of inotropes (p = 0.31) and fluid resuscitation (p = 1.11) followed by hemodynamic monitoring (p = 0.09) and difficult peripheral access (p = 0.90).

Notably, the exposed group comprised more patients with a history of prior catheterization (24.2%, n = 32) and intubation (39.3%, n = 52) in comparison to the unexposed group (catheterization: 21.6%, n = 26; intubation: 34.1%, n = 41). There was no loss follow-up (Table 1).

Table 1. Demographics of participants and an indication of CVC insertion in UCT and LCT groups

Indicators	UCT (Exposed) n=132	LCT (Unexposed) n=120	p-value
Demographics of study participants
Age a	52.64 ± 12.3 years	53.26 ± 12.3 years	0.06
Gender: ^b
1) Male	64 (48.4%)	73 (60.8%)	0.12
2) Female	68 (51.5%)	47 (39.1%)	0.08
Body mass index (kg/m)^a	24.7 ± 5.3	23.1 ± 5.9	0.08
Prior catheterisation/	32 (24.2%)	26 (21.6%)	0.12
Intubation ^b	52 (39.3%)	41 (34.1%)	0.07
Hypotension (BP <90 mmhg) ^b	111 (84.0%)	105 (87.5%)	1.00
Reason for insertion:^b
1) Delivery of inotropes	50 (37.8%)	39 (32.5%)	0.31
2) Hemodynamic monitoring	22 (16.6%)	16 (13.3%)	0.09
3) Delivery of fluids	58 (43.9%)	60 (50%)	1.00
4) Difficult peripheral access	2 (1.5%)	5 (4.1%)	0.90

Note: ^aValues are represented as Mean ± Standard Deviation, ^bValues are represented as total number (percentage)

Table 2 compares the main indicators of the UCT and the LCT groups. Access time (21.9 seconds versus 29.9 seconds), number of attempts (1.4 versus 1.8), and number of arterial pricks (n = 10 versus n = 36) were less in the UCT group with more successful first attempts (n = 82 versus n = 68) but a longer total procedural time (32.2 minutes versus 22.7 minutes). In addition, the total number of failed attempts (n = 10 versus n = 19) and the overall complication rate (4.5 % v/s 11.7%) were observed more in the LCT group. This data signifies that UCT was more beneficial in this study.

Table 2. Indicators of UCT versus LCT techniques

Indicators	Procedure Type			95% CI
	UCT, n = 132	LCT, n = 120	Difference (UCT-LCT)
Access Time (in seconds)^a	21.9 ± 8.1	29.9 ± 9.2	−8	−4.2 to −11.7
No. of Attempt^a	1.4 (0.6)	1.8 (Mittal & Coopersmith, 2014)	−0.4	−0.1 to 2.3
No. of arterial pricks^a	10 (7.6)	36 (30)	−26	−8 to −42
Total Procedure Time (in minutes)^a	32.2 (11.4)	22.7 (10.3)	9.5	8.1 to 56.2
Successful first attempt^b	82 (62.1%)	68 (56.7%)	5.4%	3.8% to 18.2%
Failure of attempts^b	10 (7.5%)	19 (15.8%)	−8.3%	−2.3% to −19.6
Overall complication rate^c	6 (4.5%)	14 (11.7%)	−7.2%	−3.3% to −11.7%

Note: ^aValues are represented as Mean ± Standard Deviation, ^bValues are represented as a total number (percentage)

Table 3 Regarding procedural complications, 4.2% of LCT procedures experienced mispositioning, whereas no mispositioning was observed in the UCT group. Relatively more insertion site hematomas were observed in the LCT group (7.5%) than in the UCT group (3.8%). No pneumothorax or hemothorax occurred in either group. However, one mortality was reported in the UCT group due to an expanding hematoma after CVC insertion.

Table 3. Procedural complications in UCT and LCT techniques

Indicators	UCT, n=132 n (%)	LCT, n=120 n (%)
Malposition	0 (0.0)	5 (4.2%)^a
Pneumothorax	0 (0.0)	0 (0.0)
Hemothorax	0 (0.0)	0 (0.0)
Insertion site hematoma	5 (3.8%)	9 (7.5%)^a
Mortality	1 (0.75%)	0 (0.0)

Note: ^ap<0.005 by using Chi-square test and Fisher’s exact test

5. Discussion

As there is not much literature available on the advantages and disadvantages of CVC insertion via UCT and LCT in an emergency setting, we have focused on evaluating the techniques in the following categories—mean access time, total procedural time, and post-procedure complications.

This prospective cohort study shows that the ultrasound-guided CVC insertion technique has a mean access time of 29.9 seconds, almost 8 seconds more than the landmark-guided technique, which had a time of 21.9 seconds. These results are not following the existing literature. One of the research articles reports that UCT requires 9.8 seconds of mean access time and 44.5 seconds in LCT, with a mean difference of 34.7 seconds (Denys & Uretsky, 1991). Another study reports 512 seconds in UCT and 115 seconds in LCT, with a mean difference of 397 seconds (Froehlich et al., 2009; Hoffman et al., 2017). Froehlich et al. concluded that the catheter placement by residents is significantly decreased when using UCT (median 919 seconds vs. 405 seconds, p 0.02) (Froehlich et al., 2009). This study shows more time consumption in administering CVC via ultra-sound guided technique. Although, this difference of 8 seconds is negligible during the patient’s clinical management.

Studies show that successful single-needle-pass venipunctures occur consistently via ultrasound guidance. The documented rate of successful first attempts in our study was 62.1% in the UCT group and 56.7% in the LCT group, similar to findings from a study by Leung et al., with 93.9% and 78.5% successful first attempts, respectively, the reason mainly difficult venous peripheral access (Leung et al., 2006; Mehta et al., 2013). The current literature supports this study as UCT guidance has been demonstrated to have better first-attempt success and overall success when compared to LCT (Gurien et al., 2018). Our data also showed that for patients whose first attempt was not achieved in a single prick, the mean total number of attempts with UCT was 1.4 and 1.8 with LCT. These results are comparable to another study that reported 1.6 total attempts with UCT and 3.5 with LCT (Filho LP et al., 2013). Additionally, UCT procedures in our study resulted in fewer failed attempts (n = 10) than in LCT procedures (n = 19). A review by Calvache et al. indicated that after three or more insertion attempts, the incidence rate of mechanical complications increases by six times (Calvache et al., 2016).

Furthermore, the total procedure time observed in the exposed group (UCT) is 32.2 minutes, and 22.7 minutes in the unexposed (LCT) group, with a mean difference of 9.5 minutes. This difference of 9.5 minutes is significant in patient management in an emergency where every second counts. To the best of our knowledge, all the previous studies have only compared access time. We could not find any study comparing total procedure time between these two techniques because, during this procedure, the emergency physician remains occupied, and time is of immense value in emergencies. We also observed that Emergency Medicine physicians preferred LCT over UCT, although UCT is the gold standard for CVC insertion (Denys & Uretsky, 1991). This preference may be due to a lack of UCT training, varying levels of operator experience, and limited availability of an ultrasound machine. Arranging the ultrasound machine and preparing it for the procedure also requires time, which may hinder using it, especially during rush hours when the ED team faces severe time constraints.

This study showed a complication rate in UCT of 4.5%, mostly from insertion site hematomas and one unfortunate mortality due to an expanding hematoma. The complication rate in LCT was 11.7%, including 4.2% of mal-positioned CVC placement and almost twice the number of insertion site hematomas than in the UCT group. This difference in complication rates may be due to more arterial pricks in LCT compared with UCT, a result also observed in a study by Denys et al (Denys & Uretsky, 1991). Overall, fewer complications were observed in UCT than LCT, a finding well-supported by already published literature (Denys & Uretsky, 1991; Leung et al., 2006).

Our study outcomes, alongside existing literature (Balls et al., 2010; Miller, 2002), suggest that UCT should be the gold standard of care in critical patients. Due to the unpredictable dynamics of Emergency Medicine, the prolonged total procedural time is a concern and must be addressed. Placing a safe and functioning CVC within the appropriate time should be prioritized.

6. Limitations

Due to time constraints and some technical issues, we could not achieve this study’s desired number of participants. The small sample size caused higher variability and resulted in the low power of the study, which will impact the true associations and generalizability of the results.

UCT is technically challenging and requires hand-to-eye coordination, a skill that develops only with practice. It is most beneficial if the operator has good practice and is already experienced in LCT. Henceforth lack of practice can be considered one of the reasons why it took total procedure time, irrespective of the additional time probe preparation and machine arrangement takes. Some critically ill patients refused to consent; therefore, the advantages of managing successful CVC insertion could not be documented in their case.

7. Conclusion

We found a significantly higher success rate with fewer procedural complications in UCT compared with LCT in the ED. However, the total procedure time was higher in UCT than in LCT, with only a minimal difference in mean access time, making UCT less attractive in the ED setting. Nonetheless, we recommend UCT as the standard of care due to its multiple benefits. It is recommended that emergency physicians should attend frequent workshops on ultrasound use to improve hand-to-eye coordination to perform UCT with ease in urgent scenarios.

Abbreviations

CVC – Venous Catheter

ED – Emergency Department

LCT – Landmark Guided CVC insertion Technique

UCT – Ultrasound Guided CVC insertion Technique

Authors’ contributions

FA& ES- Conceptualization, supervision, writing, reviewing & editing.

SS & AF- Data curation, formal analysis, writing, reviewing & editing.

UA & ZD- Methodology, writing, reviewing & editing.

Acknowledgments

Data collectors & Health Information and Management Systems (HIMS) Aga Khan University Pakistan.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Not applicable

Notes on contributors

Fareed Ahmed

Fareed Ahmed is a Senior Instructor, Department of Emergency Medicine, Aga Khan University, Karachi Pakistan.

Umaira Aftab

Umaira Aftab is a Medical Officer, Department of Emergency Medicine, Aga Khan University, Karachi Pakistan.

Salman Muhammad Soomar

Salman Muhammad Soomar is a Instructor (Research), Department of Emergency Medicine, Aga Khan University, Karachi Pakistan.

Asher Feroz

Asher Feroz is aResearch Specialist, Department of Emergency Medicine, Aga Khan University, Karachi Pakistan.

Zeyanna Dhalla

Zeyanna Dhalla is a Research Intern, Department of Emergency Medicine, Aga Khan University, Karachi Pakistan and Student MSc Health Policy and Managment at University of Michigan, Ann Arbor, USA.

Emaduddin Siddiqui

Emaduddin Siddiqui is a Associate Professor, Department of Emergency Medicine, Aga Khan University, Karachi Pakistan.