A total of 200 patients who were admitted to the emergency hospital at the Zagazig University Hospital (ZUH) from October 2021 to April 2022 were included in a prospective comparative randomized clinical study. Patients aged between 18 and 70 years of both genders who were admitted to emergency hospital units including the emergency room (ER), operation theater (OT), or emergency intensive care unit (ICU) and indicated for both elective and emergent CVC insertion were enrolled in our study. Emergency indications were defined as emergency hemodialysis, transvenous pacing, difficult peripheral intravenous (IV) line, and hypovolemic shock. On the other hand, patient refusal, patients <18 years and >70 years, platelet count < 500000/cc, INR > 1.5, PT > 15 s, PTT > 40 s (not necessarily in emergencies), thrombosis of the internal jugular vein (IJV), recent surgery, trauma or skin infection at the target insertion site, and cervical spine collar were all absolute contraindications to CVC placement into IJV and hence were excluded from the study. Randomization was done using a computer-generated randomization table in which participants were randomized in a parallel design, to either Group A or Group B in a 1:1 ratio and allocated the patients into two equal groups.

Patients were divided into two groups according to the technique of CVC insertion with 100 patients in each, Group A receiving the blind technique for CVC insertion, and Group B receiving the ultrasound-guided technique for CVC insertion. Equipment used included an Amecath triple lumen CVC kit size 7 French × 20 cm and the ultrasound machine GE Vivid E95.

Patients were assessed regarding the site of planned insertion, explanation to the patient if conscious and cooperative, and checking platelet count and coagulation profile in elective cases. Routine Regarding anticoagulant and antiplatelet therapy; routine CVC insertion procedures were considered adequate in patients on monotherapy with acetylsalicylic acid, non-steroidal anti-inflammatory drugs, or prophylactic anticoagulants (e.g., low-dose heparin, low-molecular-weight heparin, pentasackarides, thrombin inhibitors). However, in case these drugs were combined, in particular clopidogrel and acetylsalicylic acid, in addition to therapeutic anticoagulant and emergency cases; CVC insertion was performed by experienced operators using the safest possible techniques.18 Before CVC insertion, we started by insertion of a peripheral intravenous (IV) line, if possible; for the potential need in special situations when procedural sedation was necessary in semiconscious or uncooperative patients, rapid fluid administration or medication during the procedure; like antiarrhythmic medications and medications for cardiopulmonary resuscitation. Applying a full monitor (pulse oximeter, non-invasive blood pressure, and ECG) was initiated, and then the patient was positioned in a supine flat position with lower limbs slightly elevated up to 30°, a wedge was placed behind the upper back in between shoulders and the head was turned gently to the opposite side of insertion.19,20 Xie et al. and Gok et al. concluded that leg elevation for a few minutes before insertion simulates the passive leg raising test, pooling blood from lower limbs to central circulation; thus, making neck veins more engorged, dilated, and more easily cannulated making first trial more successful.19,20 Once the site was chosen, a topical antiseptic such as chlorhexidine, in case of chlorhexidine allergy we used povidone-iodine instead, was applied circularly to the skin in ever-enlarging circles under complete aseptic conditions.

All the CVC placements were performed by specifically trained 3rd-year intensive care residents under the supervision of a specific attending physician in the same room with the same ultrasound device. In group A, CVC was inserted using the blind standard technique. In group B, the ultrasound machine was employed using the linear probe, the probe was covered by a sterile cover, the ultrasound probe was placed perpendicularly making an angle that should be close to 90° with the skin, and then localization of IJV was done. The IJV was assessed for collapsibility along its whole length to ensure patency, to choose the site of the widest diameter, shallowest depth, and farthest site from the carotid artery (better not overlying) that appears non-collapsible and pulsatile, then adjusted to be the center of the ultrasound monitor view. Afterward, a local anesthetic was injected into the subcutaneous tissue at the site of planned insertion, and then the needle was carefully introduced through the skin and carefully advanced into the IJV with real-time ultrasound guidance, The needle was introduced in continuous aspiration at an angle of approximately 60° in relation to the skin in out-of-plane technique. The angle between the needle and the skin should be 30−45º in the in-plane technique.

Different US approaches were used; short-axis probe orientation and an out-of-plane view of the needle and a long-axis probe orientation and an in-plane view of the needle. In the case of short-axis/out-of-plane view, the needle appeared on the screen as a point (Fig. 1) while it appeared as an echogenic line in the case of long-axis/in-plane view. When the needle tip was seen inside the IJV, then aspiration of dark venous blood was done to confirm the needle tip position, and the syringe was removed. A guide wire was inserted through the needle into the vein, to a depth of at least 15 cm and it was visualized after introducing inside the IJV as an echogenic dot if the probe was placed out of plane or as an echogenic thread if the probe was placed in plane (Fig. 2), and finally the catheter was visualized inside the IJV.15

Figure 1.

Out-of-plane technique showing needle appearing as an echogenic dot.

Figure 2.

In-plane technique showing guidewire appearing as an echogenic thread.

Finally, it is important to point out that we planned in case of failed insertion of CVC into IJV (maximum three attempts in case of causes other than hypovolemic shock where only one attempt was allowed) due to very high collapsibility of IJV making its cannulation hard and difficult guidewire advancement, immediate shift to ipsilateral subclavian or femoral vein was done without further trials in IJV to avoid negative consequences of multiple trials like neck hematoma, carotid artery puncture, and pneumothorax.

Immediately after CVC placement, a saline flush consisting of 5 mL of normal saline was injected into the distal hub of the CVC by the interventionist while focused ultrasonography on the right atrium using the echo probe was conducted by a second resident by visualization of the heart using subcostal 4 chamber view. The appearance of opacification of the right atrium known as Rapid Atrial Swirl Sign (RASS) or Bubble Sign was judged as “immediate” (less than two seconds after injection), “delayed” (appearing more than two seconds after injection), or “absent”. Ultrasonography was performed by the resident placing the catheter and the result was recorded by another investigator. Overall, a positive RASS (negative screening test for misplacement) translated into a correctly positioned catheter, whereas a delayed or absent flush (negative RASS or positive screening test) implied a potentially misplaced catheter requiring further evaluation by chest x-ray.21

After finishing CVC insertion, chest ultrasonography was done by the intensivist to exclude the presence of pneumothorax through the presence of a lung sliding sign at the lung apex or hemothorax through assessment of the basal lung lobe in adherence to the diaphragm. This step is performed with the patient in the supine position and with a curvilinear probe. The probe was placed in a sagittal position on the anterior surface of the thorax, close to the second intercostal space, in the midclavicular line. Sonographic references include the presence of two ribs with their respective posterior acoustic shadow and the pleural line between them.16

All of the previous steps that were carried out on patients in group B in the previously mentioned sequence could be considered as the point-of-care ultrasound (POCUS) guided CVC insertion bundle; that included CVC insertion with real-time ultrasound guidance followed by confirmation of placement by application of RASS screening test and finally screening for early postprocedural complications like pneumothorax by chest ultrasound (Table 1).

The two studied groups were compared regarding patient characteristics (age, sex, and BMI), causes of emergency hospital admission, and indications for CVC insertion. Relevant clinical and laboratory data including the use of anticoagulant therapy, INR values, and platelets counts were collected in addition to systolic, and diastolic blood pressure and heart rate during insertion. Time of performance was estimated for each patient which was defined as time from skin draping to the end of fixing CVC by sutures. Post-insertion, the proper position of CVC was assessed using a chest x-ray. The incidence of complications was estimated among the two studied groups. Manifestations suggesting infection were evaluated for including fever and chills most commonly. Still, they may be masked if the patient is immunocompromised, where atypical presentations of sepsis occur including altered mental status, hypotension, lethargy, and fatigue. The exit site examination was also done to search for signs of inflammation with inspection and palpation of the subcutaneous track. Patients may report pain, swelling, or discharge from the exit site and redness surrounding or along the subcutaneous track when exit site or tunnel infections are present. If previous manifestations developed within less than seven days of insertion, it suggested early extraluminal CVC infection while it was considered late intraluminal infection if occurring after more than seven days.22 The cost of the entire procedure was calculated. The excess cost was defined as the need for more than one CVC set (price about 11 dollars), in addition to the cost of managing associated complications like intercostal chest tube insertion in case of pneumothorax, carotid artery doppler in case of accidental carotid artery puncture or antibiotics needed to treat CVC infection (average > 20 dollars is considered excess cost)21 and finally we considered the cost of post-insertion chest x-ray that was performed for all cases (price 2.8 dollars).

Written informed consent was obtained from all patients or their first-degree relatives and the study was approved by the research ethical committee of the Faculty of Medicine, Zagazig University (IRB#6827/30-3-2021) and registered on clinicaltrials.gov (NCT05338138). The work has been performed uniformly with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for studies involving humans.

All data were entered in Excel and then imported and analyzed using the Statistical Package for Social Sciences (SPSS), version 27 (IBM, Chicago, Illinois, USA). Categorical data were presented as frequencies and percentages. The chi-square test or Fisher’s exact test was used to compare the proportion between the groups. The Kolmogorov-Smirnov test was used initially to determine the normality of continuous data. Continuous data with normal distribution were described as mean ± standard deviation (SD) and compared with the independent t-test. A P-value of <0.05 indicates significant results.

This controlled clinical trial included 200 patients, whose ages ranged from 17 to 80 years with a BMI ranging from 18 to 40 kg/m2, and 56% were males. Urgent indications were found in around 22% (24% in group A and 20% in group B). No statistically significant difference was observed between the two studied groups regarding age, sex, BMI, and emergency indications (Table 2). There was no statistically significant difference between the two studied groups concerning causes of emergency hospital admission. Obstetric causes (16%) followed by respiratory failure secondary to pneumonia (12%) and traumatic brain injury (12%) were the most common causes of emergency hospital admission among group A, while traumatic brain injury (20%) followed by obstetric causes (20%) then respiratory failure secondary to pneumonia (18%) were the most common causes of emergency hospital admission among group B as shown in Fig. S1 (Supplementary material).

Regarding indications for CVC insertion, there was no statistically significant difference among the two studied groups (p-value = 0.07). In group A, the need for multiple infusions (28%) followed by a difficult peripheral intravenous line (24%), then the need for total parenteral nutrition (TPN) (16%) were the common indications while in group B, they included the need for multiple infusions (28%) and total parenteral nutrition (20%) then difficult peripheral intravenous line (16%). There were no significant differences in the number of patients receiving anticoagulant therapy, INR values, or platelet counts (P-value = 0.69, 0.7, and 0.4 respectively). Regarding the time of performance, group B had slightly but significantly less time of performance (25.7 ± 4.3; range: 18−33) compared to group A (26.9 ± 7.4; range: 15−45) (P-value < 0.001). Similarly, the mean heart rate was significantly lower in group B (101.2 ± 17.9; range: 56−126) than in group A (103.9 ± 24.5; range: 45−150) (P-value = 0.002). The details of clinical and peri-procedural data are shown in Table 3.

The proper position was confirmed using chest X-rays in both groups without significant differences (88% vs 92%; P-value = 0.34). Group B had a higher but non-significant number of patients without any complications (64% vs 52%; P-value = 0.2). However, concerning complications, carotid artery puncture with neck hematoma occurred in 8 (8%) patients in group A compared to none in group B (P-value = 0.04). While both groups had the same rate of late CVC infection, only four patients (4%) in group A experienced early CVC infection compared to none in group B (P-value = 0.04). The most common complication was failure to insert the CVC into IJV which occurred in 12 patients (12%) in group A and eight patients (8%) in group B (P-value = 0.04). Notably, in such cases, conversion into an ipsilateral subclavian vein was performed for access as a backup plan. Regarding tracheostomized patients with failed insertion into IJV. Conversion into the ipsilateral subclavian vein was also done in eight patients (8%) in group A and four patients (8%) in group B. Regarding cost, the excess cost was consumed in 36 patients (36%) in group A while all group B had usual costs (P-value = 0.001) (Table 4).

Our study is the first to evaluate the use of POCUS in CVC insertion in Egyptian hospitals. Furthermore, our study distinguishes in the comprehensive bundle that we applied in a stepwise approach that can prevent and detect any abnormality or hazard before (like IJV thrombosis), during (like carotid artery puncture), and even after the whole procedure (like pneumothorax). We also evaluated measures that were missing in other similar studies as heart rate and blood pressure monitoring during insertion, in addition to the cost-benefit of the study.

Our study faced certain limitations though. The resources utilized as a result of complications like prolonged ICU stay and for management of other complications, unfortunately, could not be assessed and were not included in the statistical analysis. In our study, we found that when CVC insertion was indicated due to hypovolemic shock; the IJV was highly collapsible even with very gentle light pressure with the ultrasound probe making the IJV cannulation and guidewire advancement a difficult and time-consuming process, so in these cases, we had to insert the CVC into the subclavian vein by blind landmark technique as it is known that subclavian vein less likely to collapse in response to hypovolemia compared to the IJV due to its proximity to the heart compared to IJV. Although this has the advantage of reducing the number of erroneous trials, this issue should be considered in real practice. This result was consistent with Killu et al who studied the internal jugular vein collapsibility index associated with hypovolemia in ICU patients.27 Regarding RASS, it is of great importance to mention that it may be delayed or absent in patients with long filling times like shock and heart failure, so alternative methods for confirming catheter placement should be considered in patients with compromised cardiac function.

Furthermore, another limitation of our study was that the pediatric patient population was excluded from our study as they needed a separate study to assess due to different body weights, neck circumference, the smaller size of veins, and the need for procedural sedation. The last limitation on clinical implication was that we neglected preprocedural cessation of antiplatelet therapy due to its long half-life, so we didn't have the luxury of time to stop for a suitable time before insertion and to assess results.