Improving Training in Nephrology Procedures: Yes We Can

Nephrology is a complex field requiring training in diverse topics, including glomerular and tubular diseases, acid-base and electrolyte disorders, dialysis, and transplantation, to mention just a few. Often lost in this academic mission is training in the procedures we perform, such as percutaneous kidney biopsy, placement of temporary dialysis catheters, and even the dialysis procedure itself. Pathophysiological characteristics and therapy of kidney diseases are covered in numerous conferences and journal clubs, whereas training for procedures usually is relegated to “see one, do one, teach one” without didactics. Assessment of competence in other areas of nephrology is rigorously tested by examinations administered by the American Board of Internal Medicine (ABIM), but procedures merely are listed as requisite, with no minimum standards for certification.1 The Accreditation Council for Graduate Medical Education (ACGME) requires that “fellows must have formal instruction, clinical experience, and must demonstrate competence in the performance of” procedures, but likewise provides no minimum standards.2 Criteria for competence in these procedures are left entirely up to the individual training programs and, as shown in a recent survey, these criteria vary widely.3

Training in all aspects of medicine unavoidably places patients at risk, but training in invasive procedures engenders particular risk. Use of simulators can minimize this risk4 and has been shown to substantially improve procedural skills in medical students and residents.5, 6 In this issue of the American Journal of Kidney Diseases, Barsuk et al7 show that this is also true for nephrology fellows, specifically for the insertion of temporary dialysis catheters. They compared the performance of “rookie” nephrology trainees from several training programs in inserting hemodialysis catheters in the right internal jugular vein before and after a formal training session by using a simulator and a simple checklist. There was substantial improvement in scores, and the score on the posttest was significantly higher than the score of a cohort of graduating (traditionally trained) nephrologists who also were tested. Although the investigators conclude that this resulted from the use of the simulator and checklist, it is likely that the traditionally trained nephrologists received little or no formal training of any type. Thus, the simpler conclusion that more training is better than less probably is more appropriate.

A particularly disturbing and alarming finding is the unacceptable performance of graduating nephrologists, with a mean score of 53% and only 1 in 6 passing. The checklist used for assessment was very basic, and anyone who cannot pass this should not be inserting catheters and certainly does not meet ABIM or ACGME requirements. The minimum passing score was 79%, but I would maintain that the checklist represents the bare minimum of required steps and the passing score should be 100%. There are a number of important items that should be added to the checklist. These include researching the vascular access history of the patient, assessing risk of bleeding and the presence of heparin-induced thrombocytopenia, visualizing the vein before starting the procedure, visualizing both vessels (if there is only one, it is likely to be the artery), compressing the vein to distinguish it from the artery and establish patency, observing pulsations and/or Valsalva maneuver to ensure there is not a proximal obstruction or stenosis, observing blood flow into the syringe for pressure and color to ensure that it is not arterial, checking for good pull-back of blood from the initial cannulation, never forcing the guidewire or dilator, and correctly interpreting the chest radiograph.

This study has several important limitations. First, the success rate of the new fellows could have been influenced by their retesting within 1 week on the same system used for training. Any training the graduating fellows might have had was not described and could have occurred several years earlier. Last, and importantly, testing was not done in a real clinical setting. Although simulation is desirable and certainly should be used when available, it is no substitute for the real thing. Many variations can occur in the clinical setting that increase the likelihood of complications and may not be re-created in simulation, such as obesity, inability to optimally position the head, presence of a tracheostomy, and difficulty passing the guidewire. Also, the simulator does not address placing catheters in the left internal jugular vein, which can be more difficult and more prone to complications. (Some, including this author, would argue that use of the left side also is associated with a greater incidence of stenosis and should be avoided.) Therefore, it would be very informative and critically important to know whether the training described in this article improved clinical performance. Despite these limitations, the investigators should be commended for introducing the concept of formal simulation-based training for nephrology procedures and documenting its feasibility and success.

Based on these findings, how should nephrology training programs proceed? I hope that the performance of the graduating nephrologists in this study is not representative, but if it is even remotely so, there is a serious problem. Deficiencies in training for nephrology procedures led the American Society of Diagnostic and Interventional Nephrology (ASDIN) to develop guidelines for training in ultrasonography, insertion of peritoneal dialysis catheters, and hemodialysis access procedures.8, 9, 10 These guidelines form the basis of a certification process for individual nephrologists and an accreditation process for training programs that has been in place since 2004. To date, more than 200 nephrologists have been certified and 7 training programs have been accredited in 1 or more areas. This process is not offered for temporary hemodialysis catheters or kidney biopsy because these procedures are covered by ABIM and ACGME requirements for certification and accreditation. Clearly, these requirements should be more rigorous. The ABIM and ACGME should look to the ASDIN for guidance given its experience and success in this area. In the absence of specific guidelines from ABIM or ACGME, training programs should adopt their own guidelines and procedures to ensure that nephrology fellows are competent, as required by ABIM and ACGME. At a minimum, fellows should undergo some formal training (with simulation if available) and undergo documented evaluation with a checklist as used by Barsuk et al.7

The study by Barsuk et al7 serves as yet another wake-up call for nephrology training programs to pay more attention to the procedural aspects of our profession.11 We need to listen and act.