Journal Logo

Technical Report

A Mixed-Reality Part-Task Trainer for Subclavian Venous Access

Robinson, Albert R. III MD; Gravenstein, Nikolaus MD; Cooper, Lou Ann PhD; Lizdas, David BSME; Luria, Isaac MSE; Lampotang, Samsun PhD

Author Information
Simulation in Healthcare: The Journal of the Society for Simulation in Healthcare: February 2014 - Volume 9 - Issue 1 - p 56-64
doi: 10.1097/SIH.0b013e31829b3fb3
  • Free


Residents from multiple disciplines routinely perform central venous line (CVL) placement. They are expected to become proficient in knowledge regarding CVL indications, contraindications and complications, and technical skills. Subclavian CVL placement training requires significant hands-on practice and repetition.1,2 Subclavian central venous access (SCVA), as part of the task of placing a central line, is associated with significant morbidities, including life-threatening adverse events, such as pneumothorax and subclavian arterial puncture. Overall, subclavian central line placement complication rates have been reported to be as high as 18%.3,4 Furthermore, there is no protocol for establishing procedural competence in SCVA. In keeping with the belief that surgical outcomes rely on sound procedural skills, the field of surgery leads the way with respect to assessment of technical skills.5 An objective method of assessing SCVA competency/skills would be desirable and a small step toward a standardized protocol.

The internal jugular (IJ), subclavian, and femoral veins are used in both adult and pediatric populations.6,7 Access of the IJ vein with ultrasound guidance has become a popular technique owing to increased success rate on first attempt and decreased complications.8,9 However, the IJ route cannot always be used or recommended, and alternative access points must be chosen. Because of lower risks of catheter-related bloodstream infections, SCVA is preferred to femoral CVA.6 In 2011, the Centers for Disease Control and the Healthcare Infection Control Practices Advisory Committee recommended that the femoral vein be avoided owing to the higher risk of catheter colonization.10 However, the frequency of SCVA is decreasing, secondary to higher rates of mechanical complication (ie, pneumothorax).7,11 Because the clavicle partially blocks the views of the subclavian vessel, it can be more challenging to use ultrasound guidance during SCVA compared with IJ CVA.12 Thus, if the skill of the operator can be improved, SCVA can become a safer option, and thus, we developed a simulator to provide this training.

Part-task trainers (PTTs) offer a promising option for teaching and acquiring clinical skills through deliberate practice. Furthermore, several studies demonstrate that simulator training enhances clinical performance13,14 and, thus, can be used to improve trainee proficiency in technical and aseptic skills for placement of central venous catheters.15–18 There has been an increased use of simulators across medical specialties to train residents before allowing them to perform procedures on patients.19–24 In the field of surgery, several virtual-reality simulators that teach clinical skills, for example, colonoscopy and laparoscopy, have been developed and validated.20,25,26 Undergoing such training before direct patient contact is advantageous for the student and, ultimately, for the patient. In the practice of anesthesiology, mannequin patient simulators (also known as full-scale simulators) with different clinical scenarios have been created to instruct and evaluate trainees without harming patients.27,28

Simulations can be physical, virtual, or mixed. An example of a physical simulator is an intubation mannequin that can be mask ventilated and/or intubated. An example of virtual simulation is a completely screen-based computer simulation, such as the Web-enabled Virtual Anesthesia Machine that is used to teach anesthesia machine function and troubleshooting.29 Virtual simulations that teach basic fiber optic intubation skills and regional anesthesia have also been created to train residents.30,31 A mixed-reality (MR) simulation or mixed simulation is, as the name implies, a hybrid version with both physical and virtual components. Many repetitions are needed for a trainee to achieve competency in performing a procedure such as subclavian venous access. Because most trainees currently experience only a few opportunities to place a subclavian CVL owing to the decreased frequency of subclavian venous access compared with ultrasound-guided IJ vein access, we created an MR SCVA placement simulator to facilitate deliberate practice.7,8 We evaluated the degree of realism of SCVA on a novel CVA mixed simulator; examined whether this PTT has the resolution to differentiate between varying degrees of experience levels in this setting, namely, expert, intermediate, and novice; and assessed whether this simulator is efficacious as an instructional tool. It was necessary to perform these evaluations before formalizing the use of the new simulator in our training program.32



The University of Florida (UF) SCVA mixed simulator is one of a series of MR PTTs.33 It is a novel PTT developed for instruction and training in SCVA (Fig. 1). The physical component of the simulator consists of a skeletal structure with bony landmarks, skin, and thorax. The physical component is a 3-dimensional (3D) buildup of a high-resolution head and thorax computed tomographic scan via a 3D printer also known as a fast prototyping machine (zPrinter 310, Z Corporation, Rock Hill, SC). The virtual components are the relevant soft tissues in the thoracic cage (lungs, subclavian vein, and artery). The 3D virtual model is registered within the physical skeletal structure, meaning that they both occupy the same 3D space because the virtual model is exactly underlaid within its corresponding physical model. The 3D physical model for the torso and neck as well as the vein, artery, and lung came from CT and magnetic resonance imaging scans, respectively, of a deidentified man. The torso, neck, and head of the simulator including anatomic landmarks such as the palpable sternal notch and the clavicle as well as selected ribs were physically modeled. We also attempted to simulate the feel of skin and underlying tissues to user touch and resistance to puncture at specific regions where the needle is usually inserted. The remainder of the simulator was virtually modeled and registered to the physical component (the torso) with submillimeter accuracy. Individual components (vein, artery, and lungs) from the magnetic resonance imaging scan were manually reconstructed into separate 3D virtual objects.

University of Florida (Gainesville, FL) SCVA simulator. A, physical model; B, virtual model and visualization (courtesy of Center for Safety, Simulation, and Advanced Learning Technologies).

The physical model is composed of a hardened powder material with soft inserts of cloth-like material representing simulated skin at the sternal notch, bilateral neck, and subclavian regions. An actual central venous syringe and needle assembly (introducer needle, 18-gauge × 2-1/2-inch XTW) (Ref Kit: CV-15802, TeleFlex, Reading, PA) is tracked in a defined 3D space via magnetic tracking (3D Guidance TrakSTAR, Ascension Technology Corp, Burlington, VT) and is used to interact with both the physical and the virtual elements of the simulator. A magnetic sensor with 6 degrees of freedom is precisely located at the needle tip. For visualization, a virtual syringe and needle assembly is represented in the 3D space based on the tracked position and orientation of the physical counterpart. The virtual syringe and needle assembly is represented relative to the virtual 3D model. Thus, if the tip of the physical needle is in the virtual subclavian vein, the user is considered to have successfully accessed the subclavian vein, and “success” is audibly announced. Conversely, if the needle tip is inside the virtual lungs or subclavian artery, a pneumothorax or arterial puncture, respectively, has occurred and is audibly announced in real time. The simulator announces “pneumothorax,” “arterial puncture,” or “backwall” in real time, so the user recognizes a complication, allowing appropriate adjustments for successful access. A database on a laptop computer that runs the simulator records the time it takes to successfully achieve SCVA for each individual user and each session. Additional recorded data are the number of attempts to successful SCVA, number of pneumothoraces, subclavian artery punctures, and “near misses” (needle tip within 5 mm of the lungs or subclavian artery). A video of the simulator can be viewed at


The study participants were anesthesiology and emergency medicine residents at postgraduate training levels postgraduate year (PGY) 1 through 5 and included faculty in anesthesiology and emergency medicine. The faculty indicated on the prequestionnaire that they had performed greater than 25 SCVA procedures; however, there was a wide range of experience among the faculty regarding how recently they had performed SCVA. The UF College of Medicine Institutional Review Board approved the study (UFIRB # 2011-U-0578). With trainees’ informed consent, 65 volunteer physicians at a single institution (11 PGY 1 clinical interns, 14 PGY 2 residents, 11 PGY 3 residents, 13 PGY 4 residents, 5 PGY 5 fellows, and 11 faculty) were enrolled from July to September 2011. Participants were also categorized as novice (PGY 1 and PGY 2), intermediate (PGY 3 and PGY 4), and expert (PGY 5 and faculty). Based on availability, residents and faculty physicians were invited to perform 3 trials on the simulator. It was made clear to the residents and faculty that they were free to refuse to participate. Standardized instructions were given to all participants during the orientation session. After orientation, participants completed (1) a prequestionnaire about previous subclavian central venous catheter placement experience; (2) simulation trial 1 (without the simulator’s visualization software visible) to establish the participant’s baseline skills; (3) a teaching intervention by the same instructor (A.R.R.), followed by simulation trial 2, with the 3D real-time visualization of the virtual lung, blood vessels, and virtual needle (Fig. 1); (4) simulation trial 3, using the simulator without the visualization software; and (5) a postintervention questionnaire about the realism and educational value of the simulator. All the simulation trials were for right SCVA, irrespective of participant handedness. An after-action review (debriefing) was held after each simulator trial. The brief teaching intervention consisted of a printed tutorial that provided a review of anatomic landmarks; point of insertion, placement, and direction of needle; and a stepwise demonstration of SCVA.1 The participants had the opportunity for practice during trial 2 and immediately after the second debriefing. The user could practice as long as desired before trial 3. All participants received the same teaching intervention, except for the faculty physicians, because a critique of faculty performance was not the objective of the study. The main test parameters were time (in seconds) to complete SCVA and the total SCVA score automatically generated by the simulator’s scoring algorithm. The SCVA score is a combination of the safety score and efficiency score. The determinants of the safety score were penalties for (1) pneumothorax or “near-miss” pneumothorax—needle tip within 5 mm of lungs; (2) subclavian arterial stick or “near miss” (arterial strike—needle tip within 5 mm of artery); (3) tracheal strike; or (4) puncture of back wall of the vein. The efficiency score was determined by the elapsed time (starting at syringe lift from the table to successful venous access or when a maximum SCVAscore of 100 had been reached), number of attempts, and number of skin punctures. The time the participant spent interacting with the simulator was documented. The participant was also penalized for time taken to complete the procedure. For each additional second taken to access the subclavian vein, an additional tenth of a point was added to the score (range, from 0 for best score to 100 for worst score). Postsimulator trial metrics were obtained, including a post–SCVA score. Additional outcome variables collected for all 3 trials were number of attempts (needle passes), number of arterial punctures, and number of pneumothoraces. All the participants’ sessions were recorded (ie, the simulator recorded the needle tip trajectory). Before using the simulator, 2 vascular surgeons not involved with the study development evaluated the simulator and judged it to be anatomically correct based on their years of placing subclavian CVLs and surgically assessing the subclavian anatomy.


The preintervention questionnaire collected demographic data (age, sex, and handedness), the experience level (training year, years in practice), and previous subclavian central venous cannulation experience (estimated number placed). In addition, the participants reported confidence in subclavian venous placement ability on an 11-point Likert scale (from 0 = not at all confident to 10 = very confident). After completing the simulation exercise, the participants completed a postintervention questionnaire. A 5-point Likert scale, varying from 1 = strongly disagree to 5 = strongly agree, was used to subjectively assess the simulator’s realism.34 The participants were also asked whether they felt that the experience on the simulator could be useful in clinical practice and whether the simulator should/could be used as a teaching/learning tool. The tactile feedback and steering of the needle were also evaluated. Participants were asked to reevaluate their confidence in SCVA after training with the simulator.

Statistical Analysis

SAS version 9.3 (SAS Institute Inc, Cary, NC) software was used to compute descriptive statistics and conduct statistical analysis of the data. A separate repeated-measures analysis of variance (ANOVA) was conducted for the outcomes of interest as follows: time to access, attempts, skin punctures, and SCVA score. Post hoc pairwise comparisons were conducted using dependent samples t tests for the within-subjects factor, trials, and independent samples t tests for the between-subjects factor, levels of training. To control the experiment-wise error rate at α =0.05, the Bonferroni correction was applied, and the P value used as the criterion for statistical significance was adjusted for the number of pairwise comparisons, for example, P = 0.0167 for trial 1 versus trial 2, trial 1 versus trial 3, and trial 2 versus trial 3. Because the responses to the survey data were nonnormal, separate Kruskal-Wallis analyses were performed to examine the effect of training level on the participants’ responses.35


Demographic variables, including age, sex, handedness, and postgraduate year of training, were obtained and are presented in Table 1. Table 1 shows that a majority of the participants were male (48/65, 73.8%) and right handed (61/65, 93.8%). The distribution of the levels of training was PGY 1 at 11/65 (16.9%), PGY 2 at 14/65 (21.5%), PGY 3 at 11/65 (16.9%), PGY 4 at 13/65 (20.0%), PGY 5 at 5/65 (7.7%), and faculty at 11/65 (16.9%). When asked to estimate the number of subclavian lines placed, 15.4% of participants reported that they had never placed one, and 4.6% had never observed this procedure. A majority of participants (66.2%) had no experience in either observing or independently performing subclavian access without assistance. Given the training levels and descriptors for experience categories, our sample was composed of a wide range of skill with this procedure.

Demographics of the Participants

If the simulator is to be used as a training tool in the development of procedural skills, we must provide evidence that task performance improves after the educational intervention. Table 2 shows the outcome measures for PGY 1 to 4 residents. From trial 1 (preintervention) to trial 3 (postintervention), the average SCVA score for resident participants (n = 49) improved by 30.3 points (from 51.5 to 21.2), and time to obtain venous access was reduced by an average of 79.4 seconds (from 117.3 to 37.9 seconds). Table 3 shows that from trial 1 (preintervention) to trial 3 (postintervention), the average SCVA score over all participants (n = 65) improved by 24.5 points (from 45.6 to 21.1), and time to obtain venous access was reduced by an average of 64.6 seconds (from 100.3 to 35.7 seconds). Complication rates for pneumothoraces and subclavian arterial punctures were reduced from 8.2% to 2.0% and 12.2% to 8.2%, respectively (Table 2). The average time residents interacted with the simulator was 8 minutes 59 seconds (minimum of 3 minutes 12 seconds and maximum of 15 minutes 56 seconds).

Summary of SCVA Access Data and Questionnaires
Quantitative Data From Simulation Trials Evaluating Time, Number of Attempts, and Number of Skin Puncture That Participants Took to Achieve Successful SCVA and the Calculated SCVA Score

Improvement in performance on key outcome measures from the 3 waves of data collection were also evaluated using repeated-measures ANOVA with follow-up pairwise dependent sample t tests. There were significant reductions in average time (F = 31.94, P < 0.0001), number of attempts (F = 10.56, P < 0.0001), skin punctures (F = 8.15, P = 0.0007), and SCVA score (F = 18.59, P < 0.0001) (Table 3). For all outcomes, follow-up tests showed significant differences between trials 1 and 2 and between trials 1 and 3 (P ≤ 0.0125), but not between trials 2 and 3. The effect size for pairwise differences, as measured by Cohen,35 ranged from 0.36 to 0.78. Although participants demonstrated increased success rates from 76.9% (trial 1) to 92.3% (trial 3), this was not statistically significant (P = 0.08).

To provide evidence for the use of the simulator as an assessment of procedural skill, there should be differences in outcomes that are dependent on experience. A split-plot ANOVA with level of training categories (novice, intermediate, and expert) as the between-subjects factor and trial (trial 1 vs. trial 3) as the within-subjects factor was conducted for the primary outcome measures, that is, time to access and SCVA score. With time to obtain access as the independent variable (Fig. 2), there were significant main effects for trial (F = 30.01, P < 0.0001) and experience (F = 4.27, P = 0.02), moderated by a significant interaction (F = 3.37, P = 0.04). For the SCVA score (Fig. 3), there was a significant main effect for trial (F = 29.68, P < 0.0001) moderated by a significant 2-way interaction between level and trial (F = 3.51, P = 0.04). The main effect of level of training was not significant (F = 2.66, P = 0.08). For both performance outcomes, before the simulation, there were significant performance differences (P < 0.05) between experts and novices and between experts and intermediates, but not between novices and intermediates. After completion of training, there were no differences in performance between levels of experience.

Subclavian central venous access score reduction before and after teaching intervention and simulation experience by subgroups (novices, PGY 1 + 2; intermediates, PGY 3 + 4; experts, PGY 5 + faculty).
Mean time to complete SCVA simulation (in seconds) at baseline and after simulator intervention by level of training. The error bars represent 95% confidence intervals. The simulator discriminated between novice and expert and intermediate and expert. CI, confidence interval.

Participants also completed a postintervention survey. There was a 100% response rate as the participants were asked to complete the survey immediately after their trials. On a 5-point scale (1 = strongly disagree to 5 = strongly agree), 90.7% of participants agreed or strongly agreed that the SCVA simulator was realistic (M = 4.3). There was also strong agreement (96.9%) that the simulator offered an accurate anatomic representation of SCVA (M = 4.7). The tactile feedback and steering of the needle was rated 4.1 and 4.5, respectively. Overall, participants were satisfied with the tactile feedback; 55 (84.6%) of 65 rated it favorably. Participants strongly agreed that the simulator should be used as a training/educational tool (M = 4.9) and strongly agreed on its clinical applicability (M = 4.9), and 95.4% endorsed the statement that experience with the SCVA simulator would help them with future SCVA. Participants expressed satisfaction (M = 4.3) with the entire training program (simulator plus teaching intervention), and 86.2% agreed that the experience with the simulator would lead to improved technical proficiency with SCVA.

Significant differences among the levels were observed only for technical proficiency (χ2 = 7.77, P = 0.003) and overall satisfaction with the simulation (χ2 = 12.16, P = 0.0005). Follow-up Wilcoxon tests show that for both items, although still very positive, the experts tended to provide lower ratings than those of both the novice and the intermediate groups.

In addition, participants rated their confidence in performing SCVA on an 11-point scale ranging from 0 = not at all confident to 10 = very confident. The mean (SD) preintervention confidence rating to perform the procedure was 4.62 (2.88) versus 7.69 (1.74) after the intervention. Results of a paired t test shows the mean difference; 3.08 was significant (t = 8.21, P < 0.0001). By convention, this represents a very large effect (Cohen’s d = 1.02).


In this study of the prototype SCVA mixed simulator, we have demonstrated evidence of its ability to help in training learners. According to the participants, on average, the simulator offers a very good representation of a real SCVA procedure. This mixed simulation approach offers a promising tool for central vascular access procedural training. The major benefit to the participants using this mixed simulator will be the opportunity to practice the SCVA procedure with no potential for patient harm. Real-patient safety and operator comfort level will likely be enhanced because trainees have more opportunity, at their convenience and on demand, to obtain SCVA experience and practice on the simulator before performing the procedure in a clinical setting. Because the participants received individual feedback on their performance, facilitated by the built-in after-action playback review feature, the mixed simulation approach provides the opportunity for deliberate practice followed by objective feedback, which is an efficacious way to become expert.36 The opportunity to practice with the simulator improved the participants’ technical skills and confidence level. In addition, the ability to document near misses is a new capability that we are not aware of in existing SCVA PTTs.

In a postsimulation survey (Table 2), faculty and fellows rated the simulator’s ability to improve technical proficiency lower than the residents. Although the faculty and fellows perceived that the simulator could improve their technical proficiency, in fact their performance remained relatively unchanged (Fig. 2). This discrepancy could be due to the expert’s interpretation that the question applied to others and not them.

Table 3 shows an increase in successful SCVA from 76.9% to 92.3%, reduction in mean (SD) number of attempts from 7.9 (7.8) to 4.6 (5.8) and mean (SD) decreased time(s) to achieve successful SCVA of 100.3 (87.7) seconds to 35.7 (42.1) seconds. Table 2 demonstrates a decreased incidence of pneumothorax and subclavian artery puncture of 6.2% and 4.0%, respectively, when the residents are analyzed alone. Our results are in agreement with previous studies that show higher success rates, fewer needle passes in obtaining CVA, and fewer arterial punctures after simulator-based training.17,18 The trainees achieved a final pneumothorax incidence of 2.0% after trial 3, which is similar to the reported rate (1.5%–3.1%) in the literature.37 The SCVA mixed simulator differs from other simulators in that potential life-threatening complications, such as pneumothoraces from failed performance and near misses, are identified, and an after-action review and automated consistent objective score are provided. Individual patterns of failure or success and trajectory are identified as each trial is recorded, and a postsimulation debriefing with replay is held, which aids in the learning process. The automated nature of the after-action review facilitates the emerging concept of self-debriefing in those instances where an expert instructor may not be available.38 The improvement in performance was only observed on the simulator. We did not seek to obtain concurrent validity, that is, does the simulator provide clinical learning that is generalizable to real-patient care settings. Recently in a systematic review, only 4 (3%) of 109 journal articles on medical simulators showed concurrent validity.39 It is possible that with our simulator, we would not see a similar reduction in errors and complications with real patients during SCVA. Further studies are warranted to determine if practice with the new simulator translates into changes in clinical practice and in improvement in patient outcome. Outcome of experts remained roughly the same from trials 1 to 3. This group was the only one that did not receive formal teaching instruction. Although not formally tested, this suggests that experts with advanced skills in SCVA may not show a significant improvement after practice on the mixed simulator. In a short training time (average time, 8 minutes 59 seconds) on the mixed simulator, novices were able to perform as efficiently (based on the SCVA score) as a faculty member experienced in SCVA.

The mixed simulator was found to be an efficacious and engaging training tool, as evidenced by the positive assessment by the subjects in the study. This study did not assess skill retention. Barsuk et al15 studied retention of central venous catheter insertion skills at 6 months up to 1 year after simulation-based mastery learning. Our study demonstrated the efficacy of a mixed PTT and showed that the participants’ performance improved and was also likely associated with a change in the subject’s confidence level and satisfaction.

Mixed simulation is ideal because it allows the learner to practice and be evaluated for clinical skills. As the shift continues toward the Accreditation Council for Graduate Education milestones, the SCVA simulator can be used to facilitate assessment and objectively provide quantifiable measures for learners across the continuum of medical education and across multiple disciplines.40,41 This evaluation tool can be used for formative or summative assessment. Alternatively, performance on the simulator could be one component of an objective structured clinical examination or an objective structured assessment of technical skills.42 The objective measure of this simulator offers a quantitative assessment guide for declaring a trainee/resident/physician technically proficient and competent in SCVA. For maintenance of certification, a practicing physician who does not perform SCVA frequently could use this new simulator to either verify competency or establish it. As a by-product of the tracking and recording of the needle tip trajectory relative to relevant anatomic structures for the SCVA technique, including the examinee’s accuracy, other factors such as precision, speed, and error rates can also be assessed.

The SCVA score is a calculated score that reflects how users would perform for actual SCVA in patients. This is a new method of analyzing SCVA generated from a simulator. The SCVA score was shown across users to have accuracy and precision in examining SCVA procedural efficiency and safety. Others researchers may chose different performance indicators for this clinical skill, but it is interesting that the experts started out doing pretty well and did not improve and that ultimately, the novices achieved the same scores as the experts. This is an encouraging indicator that the simulator is likely very similar to reality. That is, if the experts did well on the first try, they are likely using the same technique they use clinically. Because the expert scores did not change, it can be assumed that they did not learn to game the simulator to get better.

This study has several limitations. First, this study was conducted at a single institution and used only one anatomic model; thus, participants’ experience could vary if they were presented with aberrant anatomy in either a virtual-reality or physical model. Second, we had the participants assume they had aseptic technique, that is, were wearing sterile clothing (hat, mask, gown, and gloves) during each simulation. Individual performance could theoretically change if optimal sterile techniques were used. Third, we did not use a control group for the study design. Moreover, one could certainly argue that the change in SCVA score or performance was related simply to practice. Fourth, the PGY 1 participants (interns) who participated in the study had better average scores on trial 1 than PGY 2 and PGY 3 participants. This is potentially explained by the fact that PGY 1 participants had completed a recent training session in SCVA on a different physical simulator during orientation at the beginning of the academic year (July), 3 months before this study. Moreover, PGY 1 participants were scheduled for their simulation trials as opposed to the other participants’ ad hoc performances. They could have read and studied on how to do the procedure the day before in anticipation of the upcoming CVA study (attention bias). In addition, the participants’ performance was not subsequently verified on actual patients. Placing a wire into the vein and threading and securing the catheter were not simulated or evaluated. Similarly, participants were not asked to prepare the skin before attempting venous access. Thus, considering all the previously mentioned factors, this study does not totally indicate how the subject would actually perform clinically for central line placement. Although the current simulator can support a full-blown procedure, this study was focused on a PTT of SCVA and not on the procedural steps before and after it.

Mixed-reality simulation or simply mixed simulation is a rapidly growing area. Pugh et al43 demonstrated that preclinical medical students could have technical skills assessed with an ePelvis simulation. Furthermore, Deladisma et al44 showed that a single interaction with a virtual breast patient in a clinical breast examination simulator was a useful adjunct to teaching procedural skills and improving students’ confidence in breast examination and history taking. Kotranza et al45 found MR humans to be usable, acceptable, and educationally beneficial for practicing examination skills by medical students. It is a difficult challenge to establish predictive validity of a simulator. Our mixed simulator will be formally included in our training program to learn SCVA, given its decreased frequency of use and to address complications such as pneumothoraces. The mixed simulator has been disseminated to a manufacturer of CVA kits that has used it worldwide to train its customers.

An argument can be advanced that subclavian access should be performed with ultrasound, considering the recent studies noting an increase in successful placement of a subclavian vein catheter with ultrasound guidance, although, to date, ultrasound is not widely used for subclavian access.46

As the term “part-task trainer” (PTT) implies, it is a trainer to practice a part of a task, not the whole task. The PTT presented in this article does not simulate catheter insertion and placement for mainly engineering reasons. For trainers with fixed point(s) of entry, it is relatively straightforward to instrument the entry port to track depth of insertion and rotation of a catheter or trocar being inserted via the fixed entry port. In the case of training in SCVA, one of the main learning objectives was to select the correct entry point where the instrumented needle punctures the simulated physical skin.47 Because the entry point will vary, it is technically difficult to preposition sensors to track catheter placement once venous access has been obtained. Other parts of the task of central venous catheter placement such as skin prepping and draping are not simulated either. In the case of skin prepping, the challenge is not an engineering one as the developers have already developed a skin prepping simulator that could have been readily added to the SCVA simulator.48 We chose to leave out skin prepping to focus the intervention on venous access. The shorter duration without skin prepping also facilitated recruitment of clinicians as participants in the study. The current simulator can be used for IJ access as can be seen from the physical and virtual models in Figure 1. Keeping clinical realism in mind, we deliberately did not use or evaluate the simulator for training in IJ access because it is usually performed with ultrasound guidance, absent in the version that was evaluated. A proposed version of the simulator will have simulated real-time ultrasound guidance that can be used for realistic training in IJ and axillary (lateral subclavian) access.

The ability for trainees to view in real-time 3D visualizations (also visible to the instructors), instead of having to imagine them, is a distinguishing feature of the SCVA PTT. For trainees able to visualize internal anatomy that they cannot directly observe when obtaining SCVA, the MR technology may help learners in forming lifelong mental pictures of anatomically correct 3D relationships that are often otherwise opaque. Other innovative features of the CVA PTT are (a) the anatomic realism derived from the use of medical imaging of an actual human to create the 3D physical (via 3D printing) and virtual (internal soft tissues such as vein, artery and lungs) components and (b) collocating virtual and physical anatomically correct components in a mixed simulator.

Central venous access requires both cognitive and psychomotor skills. From the cognitive side, the trainee needs to have a correct 3D mental model of the anatomy surrounding the target, an appreciation of the risks and a strategy to mitigate the risks by selecting an appropriate entry point, and choosing a trajectory that minimizes collateral damage if the target is missed. Psychomotor skills such as manual dexterity and spatial ability are required to successfully place the needle tip in the vein. Generally, physical simulators are best for training in psychomotor skills, and virtual simulators, unconstrained by the laws of physics, facilitate in adding to our cognitive skills by visualizing normally invisible internal structure and anatomy.47 A mixed simulator provides the desired attributes of both physical and virtual simulators in meeting the psychomotor and cognitive learning objectives, respectively, for a CVA simulator. Trainees are physically palpating to locate the clavicle and sternal notch and are grasping and steering directly with their hands a physical needle and syringe actually used for venous access instead of using a mediated interface such as a pointing or haptic device. In addition to the virtual component providing 3D visualization, the virtual vessels do not leak after repeated “puncture” compared with tubes filled with colored liquids.

In conclusion, we have demonstrated in this study that a mixed SCVA simulator offers a realistic representation of SCVA. The simulator not only allows a high score in SCVA to be achieved but also allows the trainee to improve to a score comparable with those of experienced physicians. This portable mixed simulator has the capacity to help teach, train, and evaluate the subclavian venous access skill of clinicians across multiple disciplines. Further studies investigating this educational technology are warranted to establish validity of the mixed simulator.


The authors thank the UF Anesthesiology and Emergency Medicine residents for their voluntary time and effort with the study.


1. Braner DA, Lai S, Eman S, Tegtmeyer K. Videos in clinical medicine. Central venous catheterization—subclavian vein. N Engl J Med 2007; 357: e26.
2. Kilbourne MJ, Bochicchio GV, Scalea T Xiao Y. Avoiding common technical errors in subclavian central venous catheter placement. J Am Coll Surg 2009; 208: 104–109.
3. Merrer J, De Jonghe B, Golliot F, et al. Complications of femoral and subclavian venous catheterization in critically ill patients: a randomized controlled trial. JAMA 2001; 286: 700–707.
4. McGee DC, Gould MK. Preventing complications of central venous catheterization. N Engl J Med 2003; 348: 1123–1133.
5. Reznick RK, MacRae H. Teaching surgical skills—changes in the wind. N Engl J Med 2006; 355: 2664–2669.
6. Marik PE, Flemmer M, Harrison W. The risk of catheter-related bloodstream infection with femoral venous catheters as compared to subclavian and internal jugular venous catheters: a systematic review of the literature and meta-analysis. Crit Care Med 2012; 40 (8): 247–285.
7. Karapinar B, Cura A. Complications of central venous catheterization in critically ill children. Pediatr Int 2007; 49 (5): 595–599.
8. Troianon CA, Hartmann GS, Glas KE, et al. Guidelines for performing ultrasound guided vascular cannulation: recommendations of the American Society of Echocardiography and the Society of Cardiovascular Anesthesiologists. J Am Soc Echocardiogr 2011; 24: 1291–1318.
9. Lamperti M, Bodenham AR, Pittiruti M, et al. International evidence-based recommendations on ultrasound-guided vascular access. Intensive Care Med 2012; 38: 1105–1117.
10. O’Grady NP, Alexander M, Burns LA, et al. Guidelines for the prevention of intravascular catheter-related infections. Clin Infect Dis 2011; 52 (9): e162–e193.
11. Mansfield PF, Hohn DC, Fornage BD, et al. Complications and failures of subclavian-vein catheterization. N Engl J Med 1994; 331 (26): 1735–1738.
12. Stefanidis K, Fragou M, Pentilas N, et al. Optimization of cannula visibility during ultrasound-guided subclavian vein catheterization, via a longitudinal approach, by implementing echogenic technology. Crit Care Res Pract 2012; 617149: 1–6.
13. Anastakis DJ, Regehr G, Reznick RK, et al. Assessment of technical skills transfer from the bench training model to the human model. Am J Surg 1999; 177: 167–170.
14. Hyltander A, Liljegren E, Rhodin PH, Lonroth H. The transfer of basic skills learned in a laparoscopic simulator to the operating room. Surg Endosc 2002; 16: 1324–1328.
15. Barsuk JH, Cohen ER, McGaghie WC, Wayne DB. Long-term retention of central venous catheter insertion skills after simulation-based mastery learning. Acad Med 2010; 85 (suppl 10): S9–S12.
16. Barsuk JH, Cohen ER, Feinglass J, McCaghie WC, Wayne DB. Use of simulation-based education to reduce catheter-related bloodstream infections. Arch Intern Med 2009; 169: 1420–1423.
17. Barsuk JH, McGaghie WC, Cohen ER, Balachandran JS, Wayne DB. Use of simulation-based mastery learning to improve the quality of central venous catheter placement in a medical intensive care unit. J Hosp Med 2009; 4: 397–403.
18. Barsuk JH, McGaghie WC, Cohen ER, O’Leary KJ, Wayne DB. Simulation-based mastery learning reduces complications during central venous catheter insertion in a medical intensive care unit. Crit Care Med 2009; 37: 2697–2701.
19. Hall RE, Plant JR, Bands CJ, Wall AR, Kang J, Hall CA. Human patient simulation is effective for teaching paramedic students endotracheal intubation. Acad Emerg Med 2005; 12: 850–855.
20. Koch AD, Buzink SN, Heemskerk J, et al. Expert and construct validity of the Simbionix GI Mentor II endoscopy simulator for colonoscopy. Surg Endosc 2008; 22: 158–162.
21. Friedman Z, Siddiqui N, Katznelson R, Devito I, Davies S. Experience is not enough: repeated breaches in epidural anesthesia aseptic technique by novice operators despite improved skill. Anesthesiology 2008; 108: 914–920.
22. Blum MG, Powers TW, Sundaresan S. Bronchoscopy simulator effectively prepares junior residents to competently perform basic clinical bronchoscopy. Ann Thorac Surg 2004; 78: 287–291; discussion 287–91.
23. Wayne DB, Barsuk KH, O’Leary KJ, Fudala MJ, McGaghie WC. Mastery learning of thoracentesis skills by internal medicine residents using simulation technology and deliberate practice. J Hosp Med 2008; 3: 48–54.
24. Wayne DB, Butter J, Siddall VJ, et al. Simulation-based training of internal medicine residents in advanced cardiac life support protocols: a randomized trial. Teach Learn Med 2005; 17: 210–216.
25. Koch AD, Haringsma J, Schoon EJ, de Man RA, Kuipers EJ. A second-generation virtual reality simulator for colonoscopy: validation and initial experience. Endoscopy 2008; 40: 735–738.
26. Duffy AJ, Hogle NJ, McCarthy H, et al. Construct validity for the LAPSIM laparoscopic surgical simulator. Surg Endosc 2005; 19: 401–405.
27. Wong AK. Full scale computer simulators in anesthesia training and evaluation. Can J Anaesth 2004; 51: 455–464.
28. Lampotang S. Medium and high-integration mannequin patient simulators. In: Riley R, ed. Manual of Simulation in Healthcare. Oxford, United Kingdom: Oxford University Press; 2008.
29. Fischler IS, Kaschub CE, Lizdas DE, Lampotang S. Understanding of anesthesia machine function is enhanced with a transparent reality simulation. Simul Healthc 2008; 3: 26–32.
30. Goldmann K, Steinfeldt T. Acquisition of basic fiberoptic intubation skills with a virtual reality airway simulator. J Clin Anesth 2006; 18: 173–178.
31. Grottke O, Ntouba A, Ullrich S, et al. Virtual reality-based simulator for training in regional anaesthesia. Br J Anaesth 2009; 103: 594–600.
32. Boulet JR, Murray DJ. Simulation-based assessment in anesthesiology: requirements for practical implementation. Anesthesiology 2010; 112: 1041–1052.
33. Lampotang S, Bova FJ, Lizdas D, et al. A Subset of Mixed Simulations: Augmented Physical Simulations With Virtual Underlays. Orlando, FL: Proceedings of the Interservice/Industry Training, Simulation and Education Conference (I/ITSEC); 2012.
34. Likert R. A technique for the measurement of attitudes. Arch Psych 1932; 140: 55.
35. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Earlbaum Associates; 1988.
36. Ericsson KA. Deliberate practice and acquisition of expert performance: a general overview. Acad Emerg Med 2008; 15: 988–994.
37. Graham AS, Ozment C, Teqtmeyer K, Lai S, Graner DA. Videos in clinical medicine. Central venous catheterization. N Engl J Med 2007; 356: e21.
38. Boet S, Bould MD, Bruppacher HR, Desjardins F, Chandra DB, Naik VN. Looking in the mirror: self-debriefing versus instructor debriefing for simulated crises. Crit Care Med 2011; 39: 1377–1381.
39. Issenberg SB, McGaghie WC, Petrusa ER, Lee Gordon D, Scalese RJ. Features and uses of high-fidelity medical simulations that lead to effective learning: a BEME systematic review. Med Teach 2005; 27 (1): 10–28.
40. Green ML, Aagaard EM, Caverzagie KJ, et al. Charting the road to competence: developmental milestones for internal medicine residency training. J Grad Med Educ 2009; 1: 5–20.
41. Hicks PJ, Englander R, Schumacher DJ, et al. Pediatrics milestone project: next steps toward meaningful outcomes assessment. J Grad Med Educ 2010; 2: 577–584.
42. Sidhu RS, Grober ED, Musselman LJ, Reznick RK. Assessing competency in surgery: where to begin? Surgery 2004; 135: 6–20.
43. Pugh CM, Youngblood P. Development and validation of assessment measures for a newly developed physical examination simulator. J Am Med Inform Assoc 2002; 9: 448–460.
44. Deladisma A., Gupta M, Kotranza A, et al. A pilot study to integrate and immersive virtual patient with a breast complaint and breast examination simulator into a surgery clerkship. Am J Surg 2009; 197 (1): 102–106.
45. Kotranza A, Lok B, Deladisma A, Pugh CM, Lind DS. Mixed reality humans: evaluating behavior, usability, and acceptability. IEEE Trans Vis Comput Graph 2009; 15 (3): 369–382.
46. Orihashi K, Imai K, Sato K, Hamamoto M, Okada K, Sueda T. Extrathoracic subclavian venipuncture under ultrasound guidance. Circ J 2005; 69: 1111–1115.
47. Lampotang S, Lizdas D, Rajon D, et al. Mixed Simulators: Augmented Physical Simulators With Virtual Underlays. Orlando, FL: Proceedings of the IEEE Virtual Reality; 2013.
48. Lizdas DE, Gravenstein N, Luria I, Lampotang S. An iPad simulation of skin prepping. Simul Healthc 2012; 7 (6): 564–565.

Subclavian vein; Simulation training; Mixed-reality simulation

© 2014 Society for Simulation in Healthcare