C. Cairns1, M. Goyal1,3, J. Day1, A. Kumar4, Z. Winchester1, J. Katz1, J. Bell1, S. Fitzgibbons1,5 1Georgetown University School Of Medicine,Washington, DC, USA 3Medstar Washington Hospital Center,Emergency Medicine,Washington, DC, USA 4MedStar Health Research Institute,Hyattsville, MD, USA 5MedStar Georgetown University Hospital,General Surgery,Washington, DC, USA
Introduction: Central venous catheters (CVC) are frequently placed by resident physicians in teaching hospitals. Simulation training aims to improve their technical performance and reduce procedure-associated complications. Our objective was to determine the impact of an intense simulation-based training program on residents' rates of CVC-associated complications. We hypothesized that the CVC-associated complication rates of simulation trained residents (STRs) would be lower than those of traditionally-trained residents (TTRs).
Methods: A single center, retrospective study was undertaken at an urban tertiary care teaching hospital, evaluating all CVCs placed by residents between October 1st 2014 and January 4th 2017, following hospital-wide introduction of the novel simulation–based training program. All patients with CVCs placed by residents during the study period were included in the study. Trained investigators extracted electronic medical record data regarding resident and patient demographics, CVC type, anatomic location, and post-procedure complications. Complication rates were reported as either rate per lines placed (for immediate complications) or complication per 1000 catheter days (for delayed complications), and were compared between the two study groups using the exact Poisson test with a significant p-value set at 0.05.
Results:
During the study period, 931 CVCs were placed by residents, with the majority placed by STRs (62.3%) in the Internal Jugular (IJ) vein (74.22%). A total of 36 delayed complications, including deep vein thrombosis (DVT), pulmonary embolism (PE) and central line associated blood stream infection (CLABSI), occurred, with more delayed complications occurring at the IJ site following STR insertion (STR 4.54/1000 catheter days vs. TTR 1.51/1000 catheter days, p 0.4256). The majority of delayed complications were DVTs, with more IJ DVTs occurring after STR (n=17) as compared to TTR (n=5) placement (3.67/1000 catheter days vs. 1.08/1000 catheter days, p 0.017). There was no difference in delayed complication rates for TTRs vs. STRs at the subclavian or femoral vein sites.
There was no difference between the total mechanical complication rate (including pneumothorax, hemothorax, and arterial injury) of STRs vs. TTRs (2.6% vs. 1.7%, p .50). A total of 5 pneumothoraces occurred following CVC placement by an STR at the IJ site in comparison to 4 pneumothoraces following TTR procedures (1.0% vs. 1.31%, p 1.0). Only one pneumothorax occurred following a subclavian CVC, placed by a TTR. Only one hemothorax occurred, following an IJ CVC placement by an STR. IJ CVC catheter placement resulted in 8 arterial injuries, 7 following STR placed CVCs vs. 1 following a TTR placed CVC (1.2% vs. 0.3%, p 0.07).
Conclusion: Central venous catheters placed by simulation trained residents and traditionally trained residents have an equivalent rate of mechanical complications and a slightly increased rate of DVT following CVC placement.