Central Venous Access in Neonates: Comparison of Ultrasound-Guided Percutaneous Access and Minimal Surgical Open Methods.

Background: In neonates, percutaneous central venous catheter (CVC) insertion is often a challenging technique. Recent reports have reported the efficacy of ultrasound (US) guidance when performing such an intervention. We conducted this study to compare US-guided and minimal surgical CVC insertion regarding time and ease of insertion, reliability, and complications. Patients and
Methods: This prospective randomized study included 92 neonates scheduled for CVC insertion. They were divided into two groups: Group A (46 neonates) underwent the US-guided approach and Group B (46 neonates) underwent the surgical approach. The number of attempts and the duration of the procedure were documented in both groups. In addition, intraoperative and postoperative complications were recorded.
Results: Each of patient's age, gender, weight, and the indication of catheter insertion were statistically comparable between the two groups. The number of trials showed a significant increase in Group A (1.52 vs. 1.07 in Group Bp <0.001). Nevertheless, the time of the procedure was significantly decreased in the same group (3.68 vs. 10.21 in Group Bp <0.001). Table 2 summarizes the previous findings. Failure was encountered only in one case in Group A (2.2%), which was converted to the open surgical technique. In general, the incidence of complications showed no significant difference between the two approaches.
Conclusion: Although US-guided CVC insertion is associated with an increased number of trials, the duration of the procedure is significantly diminished with its use. Furthermore, it has a high success rate in addition to a comparable complication profile with the traditional surgical method. Copyright:

INTRODUCTION

Central venous catheter (CVC) has become mandatory in the field of neonatal intensive care. It is often used in multiple clinical situations in these populations, including fluid resuscitation and administration of drugs like antibiotics and vasoactive medications.[1] Furthermore, it may be used for hemodynamic monitoring, pacing, blood sampling, and even hemodialysis.[23] It provides relatively long-term access when there is a shortage of peripheral lines.[4]

Multiple veins are available for CVC insertion, including femoral, internal jugular, and subclavian veins.[1] Umbilical veins may also be used in newborns.[5]

Although femoral vein access is mostly used by many physicians during pediatric resuscitation, it has many disadvantages, including increased infection and thromboembolic complications.[6] In addition, the subclavian approach was associated with an 11.36% major complication rate in children <2 years old, according to a previous study.[7]

The internal jugular vein (IJV) has become the main approach for intrathoracic cannulations in pediatrics.[8] Nevertheless, it is associated with an increased risk of carotid artery puncture in pediatrics aged <5 years, especially when using the anatomical landmark technique.[9] CVC through the IJV is more difficult in neonates and infants than older children and adults.[10]

Recently, ultrasound (US) has been applied in many fields in critical care.[51112] It has been the gold standard guidance method for IJV catheterization, as it is associated with increased success and decreased complication rates.[13] It was recommended by the NICE guidelines for CVC insertion in the pediatric population.[8]

To the best of our knowledge, there is a paucity of clinical trials evaluating US-guided CVC insertion in the neonatal age group. We conducted the current study to compare both US-guided and minimal surgical CVC insertion with regard to time and ease of insertion, reliability, and complications.

PATIENTS AND METHODS

This prospective randomized trial was conducted at the Department of Anesthesia in collaboration with the Department of Pediatric Surgery at Mansoura University Children Hospital after gaining approval from the local ethical committee and Institutional Review Board of our university.

Sample size was calculated using Power analysis and sample size software program (PASS) version 15.0.5, 2017, NCSS LLC company, East Kaysville, United States for Windows (2017) using the data obtained from a pilot study conducted on 10 patients at our hospital with time to perform the procedure as the primary outcome. Patients were allocated into two groups: US Group and Open Group. Time to perform the procedure was 8.2 ± 2.79 min for the US group and 10.0 ± 1.41 for the open group. A sample size of 41 patients in each group is needed to achieve 95% power in the proposed study using two-sided two-sample unequal-variance t-test with a significance level of 5%. Ten percent drop-out is expected, so 46 patients were enrolled to each group.

We included neonates scheduled for CVC during the period between January 2016 and December 2018. This was done after gaining informed written consent from the patient guardians, following a complete explanation of the advantages and disadvantages of each approach. We excluded neonates with coagulation abnormalities or skeletal deformities.

Using the sealed envelope method, the included 92 patients were randomly divided into two groups: Group A included 46 neonates who had US-guided catheter insertion and Group B included the remaining 46 neonates who underwent surgical catheter insertion.

All cases were performed under general anesthesia in the operative theater using sevoflurane inhalation, and the airway was secured via a laryngeal mask airway.

In Group A, we used a portable US device (Arrow International®, Reading, PA, USA) with an 8 MHz transducer for viewing the neck vessels on the right side. The vein was distinguished from the nearby artery by absent pulsation and easy compressibility. After the identification of the ideal puncture site, the US probe was placed perpendicular to the right IJV, and we tried to keep the central probe mark over the center of the vein. After skin puncture with the needle, it was visualized by the US till perforating the vein wall. The catheter was inserted into the right IJV under US guidance by the Seldinger technique. Blood was aspirated into the distal syringe part to confirm the intravascular location.

In Group B (the surgical group), the right IJV was approached through a transverse incision over the sternocleidomastoid muscle, and a short segment was dissected from the nearby internal carotid artery. An appropriate size catheter was inserted into the vein by Seldinger maneuver after distal control. Finally, the catheter was exteriorized through the wound side, and the remaining skin was closed by interrupted Vicryl 4/0 sutures.

The number of attempts and the duration of the procedure were documented in both groups. The position of the catheter tip was confirmed just after its insertion and then after 6 h using a portable X-ray machine. Any intraoperative (pneumothorax, bleeding, hemothorax, air embolism, and cardiac dysrhythmia), and delayed postoperative complications (infection, delayed bleeding, leak from a tunnel, malposition, or malfunction) were recorded.

One month after catheter removal, a Doppler study was ordered to evaluate long-term vein patency.

Our primary objective was the success rate of the US, which was defined as the ability to successfully cannulate the IJV regardless of the number of trials. Our secondary objectives included the duration of the procedure, number of trials, and encountered complications.

Statistical analysis

Data collection, tabulation, and analysis were conducted by using the Statistical Package for the Social Sciences (SPSS, IBM, Inc., Chicago, Illinois, USA) version 26 for Windows. Quantitative data were tested for normality using the Kolmogrov − Smirnov test and expressed as mean ± standard deviation. Categorical data were expressed percentage and frequency. Independent sample t and Mann–Whitney tests were used for inter-group comparison of parametric and nonparametric continuous data, respectively. Chi-square test or Fisher's exact test were used for comparing two or more groups of categorical data Probability (P < 0.05) was considered to be statistically significant.

RESULTS

Patient demographics were statistically comparable between the two groups (P > 0.05). The included neonates had mean ages of 8.28 and 8.85 days, while their weight had mean values of 3095.65 and 3097.83 g in Groups A and B, respectively. Regarding gender, boys represented 58.7% and 69.6% of the included neonates, whereas the remaining patients were girls.

The indication of catheter insertion also showed no significant difference between the two groups (P = 0.399). Acute respiratory distress syndrome was present in 41.3% and 28.3% of cases, whereas septic shock was diagnosed in 30.4% and 41.3% of patients in the same groups, respectively. Other indications included meconium aspiration syndrome and congenital pneumonia. The previous data are summarized in Table 1.

Table 1

Demographic criteria and the indication of catheter insertion in the two groups

	Group A (US) (n=46), n (%)	Group B (surgery) (n=46), n (%)	95% CI	P
Age (days)	8.28±5.734	8.85±6.562	−3.1-2.0	0.661
Weight (g)	3095.65±335.630	3097.83±298.321	−133-129	0.974
Gender
Male	27 (58.7)	32 (69.6)	−0.09-0.3	0.277
Female	19 (41.3)	14 (30.4)
Cause of insertion
Meconium aspiration	9 (19.6)	7 (15.2)	-	0.399
ARDS	19 (41.3)	13 (28.3)
Septic shock	14 (30.4)	19 (41.3)
Congenital pneumonia	4 (8.7)	7 (15.2)

US=Ultrasound, CI=Confidence interval, ARDS=Acute respiratory distress syndrome

The number of trials showed a significant increase in association with the US-guided approach (1.52 vs. 1.07 in Group Bp <0.001). Nevertheless, the time of the procedure was significantly decreased in the US group (3.68 vs. 10.21 in the surgical group – P < 0.001). Table 2 summarizes the previous findings.

Table 2

Number of trials and time of the procedure (min) in the study groups

	Group A (US) (n=46)	Group B (surgery) (n=46)	95% CI	P
Number of trials	1.52±0.547	1.07±0.250	0.3,0.6	<0.001
Time of procedure (min)	3.68±0.984	10.21±2.174	−7.2−−5.8	<0.001

US=Ultrasound, CI=Confidence interval

No cases with bleeding, pneumothorax, or hemothorax were encountered in our study. However, neck hematoma occurred in 13% and 6.5% of neonates in Groups A and B, respectively. In addition, cardiac arrhythmia was detected in 21.7% and 15.2% of patients in the same groups, respectively. Infection was encountered in only one case (2.2%) in each group (P = 1).

Arterial punctures occurred only in three cases in Group A (6.5%), compared to no cases in the other group. Conversely, air embolism was encountered in 2.2% of Group B cases versus no cases in Group A.

Failure of insertion occurred only in one case (2.2%) in the US-guided group, which was converted to the surgical approach. Catheter malposition was detected in 2.2% and 4.3% of neonates on radiological assessment. After catheter removal, Doppler examination revealed vein occlusion in 4.3% and 8.7% of the studies participants. The previous data are summarized in Table 3.

Table 3

Complications in the study groups

	Group A (US) (n=46), n (%)	Group B (surgery) (n=46), n (%)	95% CI	P
Bleeding	0	0	-	1
Pneumothorax	0	0	-	1
Hemothorax	0	0	-	1
Hematoma	6 (13.0)	3 (6.5)	0.465	0.292
Arterial puncture	3 (6.5)	0	0.483	0.078
Infection	1 (2.2)	1 (2.2)	1	1
Arrythmia	10 (21.7)	7 (15.2)	0.646	0.420
Air embolism	0	1 (2.2)	2.02	0.315
Mispositioning	1 (2.2)	2 (4.3)	2.045	0.557
Failure	1 (2.2)	0	0.495	0.315
4 weeks Doppler	2 (4.3)	4 (8.7)	2.095	0.398

US=Ultrasound, CI=Confidence interval

DISCUSSION

Despite the fact that CVC insertion through the IJV has been linked to a number of problems in infants and toddlers, it is still regarded as a practical and dependable access route.[10] Although the traditional anatomical landmark technique is associated with a 90% success rate in children and adults,[14] this rate decreases down to 77%–81.3% in younger children and arterial puncture rates increase up to 11.3%–25%.[1415]

The presence of a guidance method, like the US, is crucial when cannulating neck vessels, as multiple anatomical variations in the relative positions of IJV and carotid vessels have been reported in both pediatrics[16] and adults.[1718] Anderson et al. reported an incidence rate of 18% in IJV anatomy in the included 50 children aged <6 years.[16]

Not only will the US provide information regarding the anatomical variations of neck vessels, but also it could also visualize the depth, size, and intraluminal condition of the scanned vessels before intervention.[151920]

The current study was performed to compare both US-guided and minimal surgical CVC insertion with regard to time and ease of insertion, reliability, and complications. The included neonates were divided into two groups: Group A underwent the US-guided technique and Group B underwent the open surgical approach.

One could notice the statistically comparable demographics and indication for CVC insertion between our two study groups. This poses an advantage for our study, as this ensured proper randomization method. In addition, this should also negate any bias that might have skewed the results in favor of one group rather than the other one.

We encountered failure in only one case in the US group, which makes the US-guided approach have a high success rate (97.83%). This confirms the efficacy of this guidance technique compared to surgery-guided intervention.

In a previous study that included eight neonates along with 12 older pediatric patients, Al Sofyani et al. reported a success rate of 100% for the US-guided approach, as cannulation was successfully achieved in the included 20 participants.[8] Other authors reported a 100% success rate for the US for achieved cannulation, compared to only 80% for the landmark technique.[10] In addition, other authors confirmed the 100% success rate for US guidance.[15]

Verghese et al. reported compared three approaches used for IJV cannulation in the pediatric population (the landmark, smart needle, and US-guided methods). The US achieved the highest success rate, as the three approaches had success rates of 81.3%, 77%, and 94%, respectively.[21] All of the previous studies confirmed our findings regarding the efficacy of the US in achieving IJV cannulation in the neonatal population in the intensive care unit setting.

In our study, the number of trials showed a significant increase in association with the US-guided approach (1.52 vs. 1.07 in Group Bp <0.001). This is a reasonable finding, as the vein becomes more obvious in the surgical field, with the surgical approach making it easier to cannulate. On the other hand, the application of compression, either with the probe or with the needle, may compress these small neck vessels, which in turn will need more trials to achieve success.

Other studies agreed with our trial number in the US group. Another study reported that the mean number of attempts required for IJV cannulation was 1.8 ± 1.12 and 1 ± 0.28 in the neonatal and pediatric groups, respectively (P = 0.984).[8] Chuan et al. reported that the US was superior to the landmark technique regarding the number of attempts (P = 0.006), which had mean values of 1.57 and 2.55 in the US and landmark groups, respectively.[10]

Another study reported that US guidance was associated with a significant decrease in the number of attempts (P < 0.05). The median number of attempts was 2, 2, and 1 in the landmark, smart needle, and US-guided techniques, respectively.[21] These authors did not use the open surgical method, and that could explain the superiority of the US approach.

In the current study, the time of the procedure was significantly decreased in the US group (3.68 vs. 10.21 in the surgical group– P < 0.001). Although the surgical approach was associated with fewer cannulation attempts, the time needed for skin incision, subcutaneous tissue dissection, identification of vessels, and separating a portion of IJV from the related carotid artery would wipe out this advantage.

Verghese et al. confirmed our findings regarding decreased insertion time in association with the US. They reported that the time needed for cannulation had mean values of 4.5, 8.9, and 6.6 min in the US, smart needle, and landmark techniques, respectively.[21] Another study also confirmed the superiority of the US approach compared to another method regarding cannulation time (P < 0.001). Cannulation time had mean values of 4.2 and 14 min in the US and landmark groups, respectively.[15]

Al Sofyani et al. reported that the average access time had mean values of 65.3 and 154 s in the neonatal and pediatric groups, respectively, with no significant difference between the two groups (P = 0.154).[8] Although the previous studies did not evaluate the surgical approach in IJV cannulation, they agreed with our findings regarding the superiority of the US in saving procedure time.

In the current study, we did not encounter any cases with pneumothorax or hemothorax with US guidance, and this is in accordance with Al Sofyani et al., who reported no incidence of pneumothorax or hemothorax with the use of US guidance during CVC insertion.[8] These authors reported the safety of that procedure in neonates as in older pediatric children, which confirms our findings.

In our study, arterial puncture occurred only in three cases in Group A (6.5%), compared to no cases in the other group. Verghese et al. reported a similar rate of arterial puncture with US guidance, which was encountered in 6% of the included cases.[21] Chuan et al. reported less incidence of arterial puncture with US guidance, as it was encountered in 3.1% of the included cases.[10] Another study reported no cases of arterial punctures (0%).[8]

Conversely, another study reported higher rates of arterial puncture (54%), which was more evident with the posteriorly located carotid artery, as the cannulation needle traversed the IJV in front.[18] One could see some differences between studies regarding that complication, and this could be explained by differences in the prevalence of anatomical variations, quality of US machines, and operator experience.

Our findings showed that infection was encountered in only one case (2.2%) in each group, and this lies within the range reported in the literature, which ranges between 0 and 29%.[2223]

In the current study, cardiac arrhythmia was detected in 21.7% and 15.2% of patients in Groups A and B, respectively. It is worthy of mentioning that this complication results from guide wire contact with the right atrium.[24] Another study reported that the same complication was encountered in 85 out of 201 included cases, with an incidence rate of 42%.[25] Other authors reported that atrial arrhythmias and ventricular ectopy were encountered in 41% and 25% of the included participants, respectively.[26]

Our results revealed that neck hematoma occurred in 13% and 6.5% of neonates in Groups A and B, respectively, and this is in agreement with the previous literature, which stated that the incidence of that complication ranges between 2% and 26%.[2728]

In the current study, catheter malposition was detected in 2.2% and 4.3% of neonates in Groups A and B, respectively, on radiological assessment. In an alternative study, this complication was described in about 7% of cases undergoing thoracic CVC placement.[29]

In our study, after catheter removal, Doppler examination revealed vein occlusion in 4.3% and 8.7% of the studies participants, with no statistical difference between the two groups. A previous study reported occlusion of IJV in about 10.6% of adults following neck operations.[30] According to another study, at least three-quarters of previously cannulated IJVs remain patent and can be used for another CVC insertion.[31] In fact, multiple factors are incriminated in this occlusion, including prematurity, thrombogenic criteria of the catheter material or the injected drug, long implant duration, and catheter sepsis.[3132]

All in all, we see that the application of US during CVC insertion in the neonatal age group is essential, as it decreases procedure time and complication rates. Hence, training programs should be provided for intensive care physicians and anesthesiologists to increase their experience regarding this imaging modality.

Our study has some limitations; it is a single-center study that included a relatively small sample size. Thus, more studies including more cases from different pediatric centers should be conducted soon.

CONCLUSION

Based on the previous findings, although US-guided CVC insertion is associated with an increased number of trials, the duration of the procedure is significantly diminished with its use. Furthermore, it has a high success rate in addition to a comparable complication profile with the traditional surgical method. US use should be encouraged by younger physicians who have little experience in the nonguided approaches.

Financial support and sponsorship

This study is self funding according to Mansoura University protocol.

Conflicts of interest

There are no conflicts of interest.