The internal jugular vein (IJV) is one of the preferred sites of central venous access for anaesthesiologists due to its anatomical location and characteristics.¹ Originating at the base of the sigmoid sinus, the IJV runs through the jugular foramen. It then runs parallel to the carotid artery within the carotid sheath, continuing to its most common position posterior to the sternocleidomastoid muscle and anterolateral to the common carotid artery (CCA).² It is often accessible intraoperatively and easily visualised during ultrasound-assisted central venous catheterisation.³

Complications during and after IJV access include bleeding, accidental arterial puncture, pneumothorax, malposition, thrombosis, cardiac arrhythmias and infection.⁴ Adequate training reduces complication rates,⁵ which increase with more access attempts.⁶

Point-of-care ultrasound (POCUS) gained popularity for central venous access in recent decades due to its real-time guidance capabilities, improving success rates and reducing complications.^7,8,9 The Association of Anaesthetists of Great Britain and Ireland¹ recommends ultrasound for central venous access as the standard of care.

Anatomical variations of the IJV include positional differences, absence, hypoplasia, duplication and fenestration.¹⁰ These are well-documented in various populations, including China,^11,12 Colombia,¹³ Germany,^14,15 Grenada,¹⁶ Iran,¹⁷ Italy¹⁸ and Japan.¹⁹ No study examining the African population could be identified. Anatomical variation may contribute up to 15% of complications in IJV access.²⁰ Understanding these variations is crucial for safe catheter insertion. Ultrasound aids in identifying these variations and reducing complications.^9,21,22

While extensive research exists on IJV variations, few studies have explored nor made recommendations with regard to which side of the neck is potentially a safer option for central venous access.^23,24

This was a cross-sectional descriptive analysis using ultrasound imaging done at Chris Hani Baragwanath Academic Hospital, Soweto, South Africa. Inclusion criteria included all patients aged 18 years or older presenting for elective surgery. Consecutive convenience sampling was used. Exclusion criteria included patients unable to lie supine or with a history of neck pathologies, previous neck surgeries or prior IJV access. Patients were screened at the theatre reception and recruited for the study preoperatively. Written informed consent was obtained from all individual participants involved in the study. Age, ethnicity and country of origin were noted. Patient weight was either self-reported or retrieved from the patient’s record, where available, while height and neck circumference were measured by the investigator. Body mass index (BMI) was calculated where possible.

A sample size calculation was performed based on the two-sample proportions test (Pearson’s chi-squared test) using Stata 18. The reference values were drawn from Maecken et al.,¹⁴ which had the largest sample size of the comparable studies identified: the proportions of IJV overlap with the CCA on the right (46.4%) and left (34.8%) sides served as p1 and p2, respectively. A minimum detectable difference of 15% between proportions was specified, with a two-sided significance level of 5% (α = 0.05) and 80% power (β = 0.20). This yielded a total representative sample size of 356 (178 per group).

Ultrasound examinations were standardised and performed by a POCUS-certified clinician using a GE LOGIQe ultrasound machine (GE, United States) with a 6 MHz – 14 MHz, linear probe. The same investigator performed all the scans to minimise inter-operator variability. Examinations were performed using the probe in vascular mode with the frequency set to 10 MHz. Patients were positioned supine with their heads in the neutral position. The transducer was applied to the skin with the probe marker facing the patient’s right, with the face of the probe at 45° to the sagittal axis at three levels: Sedillot’s triangle (above the clavicle) (lower), cricoid cartilage (middle) and the upper border of the thyroid cartilage (upper). This was intended to simulate procedural instrumentation while minimising increased overlap from head rotation.¹⁴ The CCA was positioned in the centre of the screen, and probe pressure on skin was kept as low as possible. If the IJV was not visible, patients were asked to perform a Valsalva manoeuvre. The IJV was identified by compressibility, but pulse wave doppler was used if the investigator was uncertain. Images were labelled and saved for later measurement and assessment.

Forty-four measurements were recorded for each patient apart from their anthropometric data. This included six sets of measurements, at each of the craniocaudal positions on both the left and the right sides, namely: IJV diameter, CCA diameter, IJV distance from skin, CCA distance from skin, distance between IJV and CCA centres, degree of overlap of the vessels (0%, < 50%, > 50%), radial relationship of the IJV to the CCA and the presence of atypical vessel variants. Internal jugular vein size is highly variable, and an amorphous shape makes measuring IJV size difficult without the usage of advanced software. The deepest vertical pool was used to determine IJV size, as this parameter is directly applicable to vascular access safety and the avoidance of posterior wall puncture. The distance from the skin to both vessels was measured as the distance to the superficial vessel wall of the main vessel body, which is also relevant for vascular access.

Radial relationship of the IJV to CCA was recorded using a clock-dial system (Figure 1). Each sector that the IJV was present in was marked and recorded. Typical position was defined as positions 1–4 on the left side of the neck and 8–11 on the right side of the neck.¹⁴ These positions were chosen to increase the typical group of patients who were likely to have an uncomplicated central venous access as there is no or minimal overlap of the IJV and CCA in these positions, as well as no posterior IJV segments.

Recruitment included 178 patients presenting to the Chris Hani Baragwanath Academic Hospital theatre complex for elective surgery; no patients met exclusion criteria. The patient characteristics are depicted in Table 1.

The principal study finding was that the IJV appeared more frequently in the typical anatomical configuration on the right side of the neck, with reduced overlap with the CCA across the upper, middle and lower positions. This observation aligns with previous research emphasising the importance of IJV anatomy in central venous catheterisation.

This study’s left versus right sided typical prevalence at the level of the cricoid cartilage (52.81% vs 70.79%) is comparable to the findings of Lin (66.4% vs 69.3%)¹¹ and Maecken (67.3% vs 76.1%).¹⁴ However, at the level of Sedillot’s triangle, this study’s incidence of left versus right typical anatomy (16.29% vs 52.81%) was significantly lower than in the studies conducted by Lim (81.8% vs 88.6%),²³ Umaña (58.9% vs 73.3%),¹³ Benter (93% vs 93%)¹⁵ and Shoja (98.2% vs 98.2%).¹⁷ Inconsistent definitions of typical anatomy and differing measurement methodologies likely underlie the wide disparity in reported observations. This disparity could also be explained by population variations. Using a clock-dial classification system permits direct left-versus-right comparison between studies. At the cricoid level, the present study recorded typical anatomy in 52.81% (left) and 70.79% (right) of cases, consistent with the findings of Maecken (67.3% and 76.1%, respectively).¹⁴

This study’s findings further support previous evidence by Gordon,²⁴ Lichtenstein²⁵ and Lim²³ that the right IJV tends to be larger in diameter than the left. The observation of greater overlap on the left also supports previous evidence by Lin¹¹ and Qin.¹² The observation that overlap increased with increasing age is supported by the studies by Troianos²⁶ and Umana,¹³ but is contrasted in the study by Shoja,¹⁷ which found decreasing overlap with increasing age.

The anatomical advantages of right sided IJV access, including larger vessel diameter, reduced overlap and more typical anterolateral position, may reduce the risk of complications such as posterior wall puncture, inadvertent arterial injury and catheter misdirection.^4,5 The lower anatomical variability on the right side may explain the higher success rates associated with right-sided central venous access noted by Sulek²⁷ and Gordon.²⁴ While this study did not directly measure clinical outcomes, these anatomical findings may provide a rationale for why right-sided access lowers complications associated with central venous cannulation.

A notable strength of this study is its bilateral, multi-level analysis of IJV–CCA relationships, which is uncommon in the existing literature. Most previous studies evaluated only a single level, typically the cricoid or Sedillot’s triangle. Only Hameedullah²⁸ and Qin¹² reported multi-level data, but limited their assessment to a single side of the neck. By providing bilateral, multi-level data, this study allows a comprehensive understanding of the anatomical course of the IJV and its positional evolution across the neck. This approach is novel and may help guide procedural planning and enhance safety in central venous access.

Additionally, this study contributes new data from an African population, a demographic absent in identified existing literature. Most prior data derive from Asian,^{11,12,17,19,28} European^{14,15,16,18,25,29} or North American populations^24,26,27,30 (Appendix 1 Table 2-A1). This sample reflects the South African demographic and adds to the global understanding of anatomical variability.

The influence of anthropometric variables was also notable. We found a weak to moderate association between higher BMI and increased IJV depth, particularly on the left side at lower levels, which could complicate cannulation. Older age was associated with deeper IJV positions on the left, especially in female patients. These observations underscore the importance of individualised vascular mapping prior to central venous access, particularly in older or obese patients.

These findings may have significant clinical implications in both high-risk patients and clinician training. In patients at high bleeding risk, such as those with borderline thrombocytopenia, the right IJV should be considered for central venous catheterisation, owing to its safer, more predictable anatomy and the associated lower complication rates.²⁷ For less experienced clinicians, right-sided access may offer a safer and more straightforward approach, reducing the likelihood of procedural errors, even though these recommendations are not present in current guidelines.³¹ Additional advantages include a more direct trajectory to the superior vena cava with reduced catheter misdirection, which further reinforces the preference for right-sided access, thereby complementing our primary anatomical findings.

This study has several strengths. Like most comparable studies, randomisation was achieved via consecutive convenience sampling, and the population examined was well-represented. This study used bilateral scanning at three levels and incorporated measurements of vessel diameter, skin depth, centre-to-centre distances and degrees of overlap, thus providing a multidimensional anatomical profile not noted in previous studies. A single POCUS-certified operator conducted all scans, reducing interobserver variability and enhancing internal consistency.

Nonetheless, certain limitations should be acknowledged. The study was performed at a single tertiary hospital and included only patients undergoing elective surgery. Thus, our findings may not apply to critically ill populations or those with altered haemodynamic states. Moreover, although inter-operator variability was minimised, replication by multiple operators may improve external validity in future research. Further studies may also explore how these anatomical differences influence procedural metrics, such as number of attempts, time to cannulation and complication rates.

This study reinforces and expands upon the established evidence that right IJV access is anatomically more predictable and potentially safer. By evaluating bilateral relationships at three neck levels, we present a comprehensive anatomical map that could inform safer practice, particularly in resource-limited settings where ultrasound may be inconsistently used. Given the variability observed, even within a single population, universal ultrasound guidance remains the only reliable safeguard against complications.