Acquiring Procedural Skills in ICUs

Copyright (c) Society of Critical Care Medicine and Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited

Critical Care Medicine www.ccmjournal.org 1

acquisition of data, drafting of the manuscript, statistical analysis, and super-vision. Drs. Thiery, Boyer, Hayon, Souweine, Papazian, Mercat, Bernardin, Combes, Chiche, Diehl, du Cheyron, L’Her, Perrotin, Schneider, Thuong, Wolff, and Zeni made substantial contributions to the intellectual content of the article in terms of acquisition of data, critical revision of the manuscript for important intellectual content, and supervision. Dr. Ricard states that he had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website (http://journals.lww.com/ccmjournal).

Dr. Boyer consulted for Coviden and received support for travel from Pfizer and MSD. His institution received grant support from Teleflex. Dr. Sou-weine received support for the development of educational presentations from Gilead and Laboratoire Français du Fractionnement et des Biotech-nologies and received support for meeting expenses from Pfizer. His insti-tution received grant support from Bayer Sante and AstraZeneca and has a patent planned for 2014. Dr. Papazian provided expert testimony for Faron, received grant support from Association Régionale d’Assistance Respira-toire à Domicile, and received support for travel from Air Liquide Santé. Dr. Mercat consulted for Faron Pharmaceuticals, received grant support from Covidien and General Electric, and received support for travel from Maquet and Covidien. His institution has patent with General Electric. Dr. L’Her and his institution have patent with Oxynov and have stock options with Oxynov (three patents are pending; Dr. L’Her is one of the founders). Dr. Wolff served as a board member for Cubist; consults for Novartis, Astel-las, and Gilead; and lectures for Novartis, Astellas, and Gilead. Dr. Drey-fuss’s institution received grant support from Pfizer. Dr. Ricard served as a board member for Covidien and received support for travel from Fisher and Paykel and lectured for Covidien. The remaining authors have disclosed that they do not have any potential conflicts of interest.

For information regarding this article, E-mail: [email protected]

Objectives: Providing appropriate training of procedural skills to residents while ensuring patient safety through trainee supervi-sion is a difficult and constant challenge. We sought to determine how effective and safe procedural skill acquisition is in French ICUs and to identify failure and complication risk factors.Design: Multicenter prospective observational study. Invasive pro-cedures performed by residents were recorded during two con-secutive semesters.

Copyright © 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins

DOI: 10.1097CCM.0000000000000049

1AP–HP, Hôpital Louis Mourier, Service de Réanimation Médico-Chirur-gicale, Colombes, France.

2UMRS-722, Univ Paris Diderot, Sorbonne Paris Cité, UMRS-722, Paris, France.

3Service de Réanimation Polyvalente, CHD Les Oudaries, La Roche sur Yon, France.

4AP–HP, Hôpital Saint-Louis, Service de Réanimation Médicale, Paris, France.

5Service de Réanimation Médicale, Hôpital Pellegrin-Tripode, CHU Bor-deaux, Bordeaux, France.

6Service de Réanimation Polyvalente, CHI Poissy-Saint-Germain en Laye, site de Saint-Germain, Saint-Germain en Laye, France.

7Service de Réanimation, CHU, Clermont Ferrand, France.8Service de Réanimation Médicale, Hôpital Sainte Marguerite, Marseille, France.

9Service de Réanimation Médicale, CHU Angers, Angers, France.10Service de Réanimation Médicale, CHU Nice, Nice, France.11AP–HP, Hôpital Pitié-Salpétrière, Service de Réanimation Médicale, Paris, France.

12AP–HP, Hôpital Cochin, Service de Réanimation Médicale, Paris, France.

13AP–HP, Hôpital Européen Georges Pompidou, Service de Réanimation Médicale, Paris, France.

14Service de Réanimation Médicale, Hôpital de la Côte de Nacre, Caen, France.

15Service de Réanimation Médicale, Hôpital de la Cavale Blanche, Brest, France.

16Service de Réanimation Médicale, CHU Tours, Tours, France.17Service de Réanimation Médicale, CHU de Strasbourg, Strasbourg, France.

18Service de Réanimation Polyvalente, Hôpital Delafontaine, Saint-Denis, France.

19AP–HP, Hôpital Bichat, Service de Réanimation Médicale, Paris, France.20Service de Réanimation Médicale et Polyvalente Hôpital Nord, CHU de Saint-Etienne, Saint-Etienne, France.

Drs. Roux, Reignier, Dreyfuss, and Ricard made substantial contributions to the intellectual content of the article in terms of conception and design,

Acquiring Procedural Skills in ICUs: A Prospective Multicenter Study

Damien Roux, MD, PhD1,2; Jean Reignier, MD, PhD3; Guillaume Thiery, MD4;

Alexandre Boyer, MD, PhD5; Jan Hayon, MD6; Bertrand Souweine, MD7; Laurent Papazian, MD, PhD8;

Alain Mercat, MD9; Gilles Bernardin, MD10; Alain Combes, MD, PhD11; Jean-Daniel Chiche, MD, PhD12;

Jean-Luc Diehl, MD13; Damien du Cheyron, MD, PhD14; Erwan L’Her, MD, PhD15;

Dominique Perrotin, MD16; Francis Schneider, MD17; Marie Thuong, MD18; Michel Wolff, MD19;

Fabrice Zeni, MD20; Didier Dreyfuss, MD1,2; Jean-Damien Ricard, MD, PhD1,2

http://journals.lww.com/ccmjournal

mailto:[email protected]


Roux et al

2 www.ccmjournal.org April2014•Volume42•Number4

Setting: Eighty-four residents.Subjects: Eighty-four residents.Intervention: None.Measurements and Main Results: Number of invasive proce-dures performed, failure and complication rates, supervision, and assistance provided. Five thousand six hundred seventeen pro-cedures were prospectively studied: 1,007 tracheal intubations, 1,272 arterial and 2,586 central venous catheter insertions, 457 fiberoptic bronchoscopies, and 295 chest tube insertions. During the semesters, residents performed a median of 10 intubations, 14 arterial catheter insertions, and 26 central venous catheter insertions. Complication rates were low, similar to those in the literature: 8.6% desaturation and 7.4% esophageal placement during intubation; 0.4% and 2.3% pneumothorax with jugular and subclavian central venous catheter insertions, respectively. We identified risk factors for failure and complications. Higher rates of failure and complications for intubation were associated with residents with no or little previous experience (p < 0.001); failure of internal jugular vein catheterization was associated with left-side insertion (p = 0.005) and absence of mechanical ventilation (p = 0.007). Supervision and assistance were more frequent at the beginning of the semester and for intubation and chest tube insertion. Finally, residents had less access to fiberoptic bron-choscopy and chest tube insertion.Conclusion: Procedural skills acquisition by residents in the ICU appears feasible and safe with complication rates comparable to what has previously been reported. We identified specific proce-dures and situations associated with higher failure and compli-cation rates that could require proactive training. Questions still remain regarding minimal numbers of procedures to attain com-petence and how best to provide procedural training. (Crit Care Med 2013; XX:00–00)Key Words: arterial catheter; central venous catheter; critical care; education; intubation invasive procedure; patient safety; resident; training

Procedural complications are a significant cause of inpa-tient morbidity and mortality (1–6). This is even more true in the ICU, where a number of these procedures

are unplanned, urgent, and performed on critically ill patients, thus increasing the risk for adverse events (7). In addition, these procedures are often performed by inexperienced phy-sicians and sometimes supervised by physicians with barely more experience (8). A conflict thus arises between balancing concerns about patient safety with the adequate and efficient training of residents. One simple means to reduce complica-tions during central venous catheter (CVC) insertion (and other invasive procedures) could be to restrict access to these procedures to experienced operators; a restriction that will eventually lead to a problematic situation when most senior and experienced physicians retire, leaving inexperienced ones in charge. This simplistic decision immediately raises sev-eral questions: when can one be considered “experienced,” how does one become “experienced,” and is experience alone a guarantee of proficiency? Indeed, if procedures have been

incorrectly taught or practiced, then experience alone is not a useful predictor of procedural competence (9). Nonetheless, certification boards require that residents achieve a minimum number of procedures during their residency to be certified. Unfortunately, little is known on how optimal and efficient these numbers are for determining competence and aiding in preventing adverse events. Actually, several studies indicate that despite these requirements, residents at the end of their training do not reach an adequate comfort level in a number of invasive procedures (10, 11). Compounding the problem, the need for critical care services is likely to increase in the coming years because of the aging population and the proportion of patients taken care of by intensivists decreasing (12). Because intensivists may graduate from various residency programs and because requirements may differ from one program to another, their competency in technical skills may also vary. To bridge potential gaps, some teams have added an extra month to their training program, with disappointing results (13). Due to the shortage of intensivists foreseen in the United States by the Health Resources and Services Administration, calls have been made for a common curriculum for critical care with cross-specialty training, as exists in France and other parts of Europe (14).

Residents in France may access the intensive care medi-cine specialty via several primary specialties (anesthesiology, pneumology, cardiology, nephrology, and internal medicine). Specialty training is divided into 6-month semesters (a “sum-mer” semester, from May 1st to October 31st, and a “winter” one, from November 1st to April 30th), with residents training in a single ward (or department) for a semester and then going onto another ward or department. The work force model in French university hospital ICUs is a closed ICU with board-certified physicians. At night, a qualified attending physician is present in-house at all hours with a resident. To obtain quali-fication in intensive care, residents have to complete, in addi-tion to their primary specialization, at least two semesters in a medical ICU during their residency and then two semesters during their fellowship. There is, to date, no specific number of procedures a resident needs to perform to get his or her quali-fication. Two of the difficulties in setting a prerequisite num-ber of required procedures are to ensure that 1) this number brings trainees to a level of comfort and a zone of safety or competence and 2) residents have access to enough opportu-nities to perform these procedures. Because data in this field are still lacking in the United States (8) and scarce in Europe, we designed a prospective multicenter study to answer the fol-lowing questions: how many procedures do residents perform during a semester, and under what kind of supervision? What are the failure and complication rates of the procedures and their associated risk factors?

METHODSA multicenter observational study was conducted in 18 ICUs in teaching hospitals throughout France, during two consecutive semesters, a summer one and a winter one. Participating ICUs were contacted through the Collège National des Enseignants


Clinical Investigation


de Réanimation (CNER), French National College of Intensive Care Medicine.

Residents were asked to prospectively fill every invasive procedure they performed in a physical logbook. The log-book was elaborated through consensual item generation and pretested in two ICUs (Colombes and La Roche-sur-Yon) to ensure its feasibility and appropriateness. Logbooks that con-tained no procedure for more than three consecutive months were excluded. To ensure exhaustive tracking of procedures in the logbooks, one author (D.R.) directly contacted residents in participating ICUs by phone several times during the semes-ter to check that they were actively reporting every procedure. ICU directors or their deputy were also contacted (by J.-D.R.) and reminded to encourage their residents to complete log-books. At the end of each semester, anonymous logbooks were centralized for analysis.

The following items were collected: age, primary specialty, and previous experience in invasive procedures with the fol-lowing arbitrary stratifications: class 0 (procedure never performed), class 1 (between 1 and 10 procedures), class 2 (between 11 and 20 procedures), class 3 (between 21 and 50 procedures), and class 4 (above 50 procedures).

For each procedure performed, the following items needed to be addressed by the resident: type of procedure, date, time (day or night shift), degree of urgency (planned or emergent), presence of invasive mechanical ventilation (which required initiation or escalation of IV sedation and analgesia), severe hypotension/shock (defined as systolic arterial blood pres-sure below 80 mm Hg), infusion of vasoactive drugs or ongo-ing renal replacement therapy, indication for bronchoscopy (microbiological sampling of the lung, bronchoaspiration, or diagnostic procedure), presence of a senior doctor (a distinc-tion was made between oversight supervision and direct help), and number of attempts before success or failure. For each type of procedure, every procedure was numbered in a chronologi-cal order for each individual resident participant to evaluate the effect of experience gained over time; this effect will be called “chronological rank.”

Regarding complications, we deliberately focused on per-procedure complications (i.e., those that could be directly related to the resident’s act). The following procedure-specific complications were recorded: for intubation: arterial oxygen desaturation (defined as > 20% drop in Spo

2), aspiration

(either visualized during the procedure or diagnosed on chest radiograph commonly performed within 1 hr following intu-bation), esophageal intubation, dental injury; for CVC inser-tion: arterial puncture, pneumothorax and severe hematomas (requiring surgery and/or RBC transfusion; this was also the case for arterial catheterization); for bronchoscopy: desatura-tion (same definition as during intubation), aspiration, respi-ratory distress requiring noninvasive ventilation or intubation, hemorrhage due to bronchoscopy (macroscopic visualization during the procedure of blood due to mucosal injury by the bronchoscope); and for chest tube insertion: hemorrhage, intraparenchymal localization, or pulmonary breach.

This pure observational study was approved by the CNER and the French Society of Intensive Care. Informed consent of patients was not required for this observational study accord-ing to the French law on biomedical research.

Statistical analysis was performed with SPSS Statistics 13 (SPSS Inc, Chicago, IL). Data are presented as median (25–75% interquartile range). Univariate and multivariate logis-tic regressions were used to identify factors associated with either failure or complications among each type of proce-dure. Variables yielding p values of less than 0.10 by univariate analysis and considered clinically relevant were entered into a forward multivariate logistic regression analysis with a critical removal p value of 0.1. Odds ratios (OR) and their 95% CIs were computed. Interactions were tested. To assess progression in procedural skill competence during the semester (hypoth-esizing that the first procedures performed during the semester would be associated with more failures or complications than the later ones), chronological rank was separately analyzed for less experienced residents (classes 0, 1, and 2). A p value of less than 0.05 was considered statistically significant.

RESULTSAll the 92 residents in the 18 participating ICUs agreed to take part in the study. During both semesters, 92 logbooks were completed. Eight were excluded because they met our exclu-sion criteria.

Study PopulationThe median [25-75% interquartile range] age of the 84 res-idents was 27 years (26–28 yr), and the male/female ratio was 1.5. The primary specialties represented were cardiology (26 of 84, 31%), anesthesiology (24 of 84, 28.5%), and other medical specialties (24 of 84, 28.5% including internal med-icine, pneumology, nephrology, and family medicine [10 of 84, 12%]). The median number of previous completed semesters as residents was 5 (3–6). Previous experience in invasive procedures is detailed in Figure 1. As a whole, for arterial and CVC insertion and for intubation, residents could be equally divided into class 0 (never performed): 33%; class 1: 33%; and classes 2, 3, and 4: 33% with the exception of intubation, for which 30% of residents were in class 4. For fiberoptic bronchoscopy and chest tube insertion, a majority of the residents had no experience.

Invasive Procedures Performed by ResidentsDuring the two semesters, 5,617 invasive procedures were studied: 1,007 tracheal intubations, 1,272 arterial catheters (862 radial and 397 femoral with site of cannulation not pro-vided in 13 instances), 2,586 CVCs (1,065 internal jugular, 621 subclavian, and 900 femoral), 457 fiberoptic bronchoscopies, and 295 chest tube insertions. Figure 2 shows the median number of procedures performed per resident during a semes-ter. The number of each procedure per resident was similar in the two semesters (Fig. S1, Supplemental Digital Content 1, http://links.lww.com/CCM/A786). Fiberoptic bronchoscopy and chest tube insertion were the only two procedures that

http://links.lww.com/CCM/A786


Roux et al


were sometimes never performed during the study: 44% of residents (37 of 84) did not have the opportunity to perform fiberoptic bronchoscopy. Seventeen percent of residents (14 of 84) never inserted a chest tube and 52% (44 of 84) inserted less than five. Figure S2 (Supplemental Digital Content 1, http://links.lww.com/CCM/A786) shows, for each ICUs, the mean numbers of each invasive procedures performed by residents during a semester.

Failure and Complication Rates of Invasive ProceduresTable 1 details failure and complication rates for each proce-dure. Rates of pneumothorax during CVC insertion were 0.4% for the internal jugular vein and 2.3% for the subclavian vein. Concerning arterial catheterization, neither distal necrosis nor hematoma requiring RBC transfusion or surgery was reported. Complication rates were 17.5%, 5.7%, and 9.9% for intubation, fiberoptic bronchoscopy, and chest tube insertion, respectively.

Risk Factors for Failure of Invasive ProceduresTable 2 details risk factors for CVC insertion failure. Greater experience was associated with lower failure rate for femoral

Figure 1. Previous experience in invasive procedures. Bars represent the percentage of residents in each classes of experience for all invasive procedures. Class 0 = never performed, class 1 = performed one to 10 times, class 2 = 11–20 times, class 3 = 21–50 times, and class 4 = more than 50 times. Roughly, one third of the residents never performed any arterial/venous catheterization or tracheal intubation, one third performed between 1 and 10 times these procedures, and the last third performed more than 20 times before the inclusion in the study. More than half of the residents did not have any experience in bronchoscopy or chest tube insertion. CVC = central venous catheter.

Figure 2. Median number of invasive procedures. The numbers of invasive procedures performed by a resident during the semester in an ICU are represented as median and interquartile for each individual procedure. CVC = central venous catheter.

TabLE 1. Failure and Complication Rates (%)

Procedure Failure Complication

Central venous catheter insertion

Internal jugular vein (n = 1,065) 13.7 9.1

Pneumothorax 0.4

Arterial puncture 8.7

Subclavian vein (n = 621) 13.8 7.0

Pneumothorax 2.3

Arterial puncture 4.7

Femoral vein (n = 900) 9.2 8.3

Arterial catheter placement (n = 1,272)

14.0 Not reported

Tracheal intubation (n = 1,007) 17.6 17.5

Desaturation 8.6

Aspiration 2.6

Esophageal intubation 7.4

Dental injury 0.5

Cuff malfunction 1.1

Fiberoptic bronchoscopy (n = 457) 1.1 5.7

Desaturation 3.2

Need for noninvasive ventilation or intubation

0.2

Aspiration 3.2

Bleeding 0.7

Other 0.9

Chest tube insertion (n = 295) 4.5 9.9

Bleeding 3.1

Intraparenchymal position 0.7

Pulmonary breach 0.7

Misplacement 3.0

Othera 4.5aOccurrence or increase of a pneumothorax, liver or kidney injury, or chest tube accidentally removed during the procedure.






and internal jugular vein catheterization (OR [per experience class]: 0.78 and 0.81, respectively, all p < 0.01). This was not the case for subclavian CVC insertion. Presence of mechanical ventilation and left-sided vein catheterization were associated with internal jugular vein catheterization failure rate (OR, 0.56 and 1.76, respectively, all p < 0.01).

Regarding arterial catheterization, the presence of vasoac-tive drug infusion negatively affected catheter insertion (OR, 1.42; 95% CI, 1.016–1.989; p < 0.04).

Table 3 shows the risk factors associated with tracheal intu-bation failure. An interaction between previous experience and degree of urgency was found. The analysis was thus repeated separating inexperienced residents (classes 0–2) and experi-enced ones (classes 3 and 4). Intubation failure was associated

with degree of urgency and chronological rank (OR 3.83, p < 0.01; 0.92, p < 0.001, respectively) (Table 3) for the least experienced residents. Risk factors were not identified in expe-rienced residents (classes 3 and 4).

Chronological rank was associated with chest tube insertion failure in univariate analysis but not in multivariate analysis (Table 4). No other risk factor was identified.

Risk Factors for Complications of Invasive ProceduresThe number of attempts was the only variable associated with a higher rate of complications for CVC insertion common to the three sites (femoral OR [per attempt]: 1.19; 95% CI, 1.06–1.33; p < 0.01; internal jugular OR: 1.70; 95% CI, 1.48–1.94; p < 0.001;

TabLE 2. Risk Factors for Central Venous Catheter Insertion Failure

Variables

Univariate analysis Multivariate analysis

Odds Ratio (95% CI) p Odds Ratio (95% CI) p

Femoral vein

Experience (per class) 0.77 (0.63–0.93) 0.007 0.78 (0.64–0.95) 0.01

Chronological rank 0.95 (0.91–1.00) 0.06 0.96 (0.91–1.01) 0.09

Night shift 1.01 (0.61–1.71) 0.96 —

Left side 0.97 (0.60–1.56) 0.89 —

Urgent 1.19 (0.74–1.89) 0.47 —

Mechanical ventilation 0.73 (0.46–1.18) 0.20 —

Shock 1.22 (0.77–1.94) 0.39 —

Renal replacement 1.04 (0.65–1.67) 0.87 —

Internal jugular vein

Experience (per class) 0.83 (0.71–0.96) 0.01 0.81 (0.69–0.94) 0.006

Chronological rank 0.99 (0.96–1.02) 0.40 —

Night shift 1.34 (0.89–2.04) 0.16 —

Left side 1.70 (1.15–2.49) 0.007 1.76 (1.19–2.60) 0.005

Urgent 0.92 (0.65–1.32) 0.65 —

Mechanical ventilation 0.59 (0.39–0.89) 0.01 0.56 (0.37–0.86) 0.007

Shock 0.87 (0.61–1.25) 0.45 —


Subclavian vein

Experience (per class) 0.88 (0.73–1.07) 0.21 —


Night shift 1.23 (0.66–2.28) 0.52 —

Left side 1.15 (0.72–1.82) 0.56 —

Urgent 0.86 (0.51–1.47) 0.59 —

Mechanical ventilation 1.00 (0.59–1.71) 0.99 —

Shock 1.30 (0.79–2.13) 0.29 —


Dash indicates that these variables were not included in the multivariate analysis as the p value in the univariate analysis was not significant.

Copyright (c) Society of Critical Care Medicine and Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited Copyright (c) Society of Critical Care Medicine and Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited

Roux et al


and subclavian OR: 2.02; 95% CI, 1.61–2.54; p < 0.01). Greater experience was associated with higher rate of complications during subclavian vein catheterization (OR, 1.46; 95% CI, 1.12–1.89; p < 0.01). Complications during tracheal intubation (Table 5) were associated with lower experience levels and degree of urgency (OR, 0.82; p < 0.001 and 1.98, p < 0.05, respectively). No interaction was observed between these two factors, which

means that urgent procedures were a risk factor for complica-tions regardless of the experience level of residents.

Greater experience was also associated with a higher rate of complications during chest tube insertion (OR, 1.75; 95% CI, 1.16–2.64; p < 0.01). Experience was associated with fewer complications during fiberoptic bronchoscopy (OR, 0.89; 95% CI, 0.81–0.98; p < 0.03).

TabLE 3. Risk Factors for Tracheal Intubation Failure and Failure With Less Experienced Residents (Classes 0–2)

Variables



Tracheal intubation failure

Experience (per class) 0.70 (0.62–0.79) < 0.001 Interaction

Chronological rank 0.93 (0.90–0.96) < 0.001

Night shift 1.27 (0.87–1.85) 0.21

Nasal 1.33 (0.67–2.65) 0.42

Urgent 2.30 (1.31–4.02) 0.004 Interaction

Tracheal intubation failure with less experienced residents

Chronological rank 0.92 (0.88–0.95) < 0.001 0.92 (0.88–0.96) < 0.001

Night shift 1.46 (0.95–2.24) 0.09 —

Nasal 1.08 (0.45–2.59) 0.86 —

Urgent 2.85 (1.48–5.49) 0.002 3.83 (1.70–8.61) 0.001


TabLE 4. Risk Factors for Chest Tube Insertion Failure

Variables



Experience (per class) 1.23 (0.66–2.29) 0.52 —


Night shift 2.99 (0.88–10.22) 0.08 3.11 (0.89–10.78) 0.07

Urgent 0.41 (0.11–1.52) 0.18 —


TabLE 5. Risk Factors for Complications During Tracheal Intubation

Variables



Experience (per class) 0.82 (0.74–0.92) 0.001 0.82 (0.73–0.96) < 0.001

Chronological rank 0.98 (0.96–1.01) 0.26 —

Night shift 1.00 (0.68–1.46) 0.99 —

Nasal 1.04 (0.50–2.19) 0.91 —

Urgent 1.93 (1.13–3.29) 0.01 1.98 (1.15–1.38) 0.013





Improvement During the SemesterChronological rank (reflective of advancing experience) was sig-nificantly associated with decreased failure for CVC insertion, arterial catheterization, and tracheal intubation for residents with no or little previous experience (classes 0–2) (Table 6). Complications were not affected by the chronological rank except for fiberoptic bronchoscopy (Table 6).

Supervision and HelpApproximately one third of all procedures (32.7%) were supervised with wide differences among procedures as shown in Table 7. Supervision was provided much less often for cath-eter insertion than tracheal intubation, bronchoscopy, or chest tube insertion. Among catheters, subclavian vein cannulation required the most supervision and help. Attending physician assistance was required for 22.7% of all procedures. Venous and arterial insertions required less help than the other pro-cedures. Supervision was significantly associated with lack of experience (per class) and chronological rank. This was true for all procedures (for arterial and venous catheterization, OR ranged between 0.59 and 0.88; all with p < 0.001; for the other procedures, OR ranged between 0.41 and 0.98, all with p < 0.02), except for chest tube insertion, for which chrono-logical rank was not associated with supervision. In addition, night shift was also associated with lower supervision for tra-cheal intubation (OR, 0.63; 95% CI, 0.44–0.91; p = 0.013)

and fiberoptic bronchoscopy (OR, 0.41; 95% CI, 0.21–0.83; p = 0.012). In multivariate logistic regressions, urgency was not associated with supervision for any of the procedures.

DISCUSSIONThis is the first prospective multicenter study on invasive pro-cedures performed by residents that encompasses most techni-cal skills required to manage critically ill patients in the setting of a closed ICU system with fulltime ICU staffing and cross-specialty trainees. Procedural skills acquisition by residents in the ICU appears feasible and safe with complication rates com-parable to what has previously been reported. We were able to identify specific procedures and situations associated with higher failure and complication rates that could require proac-tive training. Taking all of this into account may help to bet-ter determine the balance between patient safety, quality, and education.

Concerns have been raised regarding the minimal number of procedures that residents had to perform, either because residents were not able to achieve them or because despite hav-ing reached them, residents still did not feel confident (10, 11). If these numbers are to be operative, they must truly reflect a step in the trainees’ progression after which residents feel com-fortable and confident. This must also translate to greater pro-ficiency, that is, fewer failures and complications.

TabLE 6. Impact of Chronological Rank on Failure and Complication Rates in Least Experienced (Class 0–2) Residents

Procedures

Failure Complications


Central venous catheter insertion 0.97 (0.95–0.99) 0.01 0.99 (0.97–1.02) 0.68

Arterial catheterization 0.97 (0.94–0.99) 0.03 Not reported

Tracheal intubation 0.92 (0.88–0.95) < 0.001 0.97 (0.94–1.01) 0.12

Chest tube insertion 1.24 (0.99–1.53) 0.05 1.15 (0.98–1.36) 0.09

Bronchoscopy 0.71 (0.48–1.06) 0.09 0.88 (0.79–0.99) 0.03

TabLE 7. Rates of Urgent, Supervised, and Helped Procedures

Procedures Urgent (%) Supervision (%) Help (%)

Central venous catheter placement

Internal jugular vein 47.7 20.8 19.7

Subclavian vein 27.1 25.7 22.2

Femoral vein 53.1 17.6 16.9

Chi2, p < 0.001 < 0.001 < 0.05

Arterial catheterization 47.7 10.0 11.2

Tracheal intubation 83.7 82.7 32.5

Fiberoptic bronchoscopy Not reported 37.4 32.8

Chest tube insertion 41.7 54.2 35.7

Copyright (c) Society of Critical Care Medicine and Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited Copyright (c) Society of Critical Care Medicine and Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited

Roux et al


There are no official guidelines as to how and when pro-cedural skills should be taught in the ICU. Patients in French teaching hospital ICUs are taken care of by residents under the supervision of a fellow and/or an attending physician. This means that a priori the resident will perform the invasive procedure his/her patient requires. It is the current practice to seek out a resident’s experience with the procedure before-hand and, depending on how comfortable and confident he/she feels (although this perception may not equal real-ity), to let the resident proceed with appropriate supervision. The number of procedures a resident will be able to perform depends on multiple factors many of which are difficult to control for. Patient case mix, patient-to-resident ratio, and ICU staffing are intuitively the most important ones. Figure 2 indi-cates that residents in our study were able to insert a median of 26 CVC, 14 arterial catheters, 10 endotracheal tubes, and three chest tubes and perform one bronchoscopy during a semes-ter. These figures are important to know if one decides to set a minimum of procedures for certification. However, we do not yet know from our study what these minimal numbers should be in order to achieve true competence and how these num-bers are learner dependent, further challenging supervisors to individualize training efforts. Our results seem to indicate that residents may not have the opportunity to perform enough invasive procedures to reach the comfort zone, an observation also made with U.S. residents (10, 11). This is particularly the case for tracheal intubation, for which a minimal number of 57 is found in the literature (15). Other opportunities must therefore be found to practice intubation. These may include practicing intubation in the operating room and participating in simulation-based training sessions (16). One could question placing fiberoptic bronchoscopy in the technical curricula of ICU residents. Several arguments, however, call for its incor-poration. Bronchoscopy appears in most difficult airway man-agement algorithms, is often required in ventilated patients, and is a procedure residents should acquire according to the European CoBaTrICE’s core competencies (17). We have to acknowledge, however, that 44% of our residents (37 of 84) never had the opportunity to perform this technique (Fig. S2, Supplemental Digital Content 1, http://links.lww.com/CCM/A786), the main reason being that in some ICUs, bronchos-copy is not performed by the ICU staff but by pneumologists. Considerable efforts should thus be made to increase the num-ber of residents trained in this procedure.

How Did the Training of Our Residents in Procedural Skills Compare With the Literature?In one of the first studies on the subject, Sznajder et al (18) reported a 15% and 17% failure rate for subclavian and inter-nal jugular vein cannulation, respectively. Our 13.7% and 13.8% failure rates, respectively, are in agreement with these figures. Schummer et al (19) recalled that even in experienced hands, successful insertion is not 100%. This important fact should be recalled to residents to reduce their potential guilt in case of fail-ure. Not surprisingly, we found that less experience with most procedures was associated with increased failure rate. This was

not, however, the case for subclavian vein cannulation, arterial catheterization, and chest tube insertion. The finding regarding the subclavian vein and chest tube insertion may seem coun-terintuitive. However, one may hypothesize that residents with a greater experience (but insufficient to be competent) were overconfident (2) or were assigned more difficult patients (15). Regarding arterial catheterization, the most likely explanation is that patient condition had a greater impact on failure than for CVC insertion. We found little data to compare ours with, regarding chest tube insertion. One study reported that thora-centesis was the procedure with which residents felt the least confident (11). This indirectly agrees with the fact that our resi-dents did not have much opportunity to perform this procedure (a median of five per resident but 17% [14 of 84] did not insert any chest tubes during the study). For this procedure, as for fiber-optic bronchoscopy mentioned above, training should probably not stop once the residency program is completed but should continue during fellowship. Our 17.6% intubation failure rate is consistent with the 20% rate reported in first year postgraduate emergency medicine residents (20). One must bear in mind that, as a whole, difficult intubations are far more frequent in the ICU than in the operating room (21), and this may not constitute the ideal setting to train inexperienced residents.

Our rates of complications for the different procedures are comparable to those reported in the literature. More specifi-cally, rates of pneumothorax associated with CVC insertion in our study are in the range reported in the literature (internal jugular vein: 0.4% vs 0.2–0.8% and subclavian vein: 2.3% vs 1.5–3.1%) (22). Similarly, our rate of complications dur-ing tracheal intubation is also in the range of those reported by others. Our 8.6% desaturation compares closely with 9% reported by Ricard-Hibon et al (23). We found a 7.4% esopha-geal intubation prevalence, in agreement with Adnet et al (24) (6.4%) and Schwartz et al (25) (8%). Our 2.6% aspiration rate is consistent with Jaber et al (21) (2%).

As indicated earlier, one of the major difficulties in technical skill training resides in finding the necessary balance between quality, patient safety and education (26), and conciliating clinical with educational priorities. Striking such balance in the ICU is even more strenuous given the fragility of the patients and the urgency of some procedures. Our complication rates in the range of those reported in the literature suggest that train-ing of our residents in procedural skills was not associated with an excess of complications. One could argue that only the “easy” procedures were offered to the residents. We believe that with an average of 40% of the studied procedures being “urgent” in nature, this was not the case. In addition, figures coming from studies performed in French ICUs indicate that residents per-form a majority of invasive procedures, for example, 75% fem-oral catheters (27) and 59% tracheal intubations (21). Even a “difficult” and potentially dangerous venous access like the sub-clavian vein was performed in 42% of cases by residents (28).

SupervisionNot surprisingly, less experienced residents were more often supervised. This supervision did not, however, prevent






complications. In the United States, residents have complained that in some instances, supervisors were not much more expe-rienced than them (8). As such, efforts should be made to improve supervision by fostering research focusing on how to optimize and standardize it (29).

LimitationsOur study has obvious limitations. By design, the study relied on residents’ self-reporting of performed procedures, failures and complications, inherently introducing recall, and reporting biases. As indicated above, various failure and complication rates are consistent with the literature guaran-teeing some external validity. Similarly, comparison between the two semesters ensured internal validity. We were not able to track how many procedures were not performed a priori by the residents either because they did not feel confident or because the attending physician preferred to perform the procedure him/herself. However, our study reflects the actual practice within French ICUs in which training resi-dents perform a majority of procedures. Finally, real-time ultrasound-guided vein cannulation was not systematically used by residents in all ICUs and not reported in the log-book. However, because studies have clearly shown its ben-efits, residents should be trained and encouraged to use this method (30).

Implications for acquiring Procedural SkillsIt is clear that the aphorism “see one, do one, teach one” cannot survive compelling data on learning curves (26, 31). Nonetheless, teaching such skills is constrained by balanc-ing patient safety with the professional obligation to teach and train with enough opportunities for residents to perform procedures and gain confidence and competence. We have identified situations at risk of failure or complications (for instance, number of attempts for CVC, urgent intubation, and left-sided internal jugular vein catheterization). Residents and supervisors should be informed of these risk factors. New educational tools such as simulation should be used to better prepare residents to perform procedures in such situations. Our study indicates that in a country in which a shortage of ICU work force is not yet present, teaching procedural skills can be significantly improved. In countries that already face such a shortage, ensuring adequate teaching of invasive proce-dures may be difficult now and even harder in the near future. Future research will have to identify effective means to help overcome anticipated difficulties in procedural skill acquisi-tion by residents.

aCKNOWLEDGMENTWe thank all the residents who participated in the study for their efforts in tracking the procedures in the logbook.

REFERENCES 1. Brennan TA, Leape LL, Laird NM, et al: Incidence of adverse events

and negligence in hospitalized patients. Results of the Harvard Medical Practice Study I. N Engl J Med 1991; 324:370–376

2. Ayas NT, Norena M, Wong H, et al: Pneumothorax after insertion of cen-tral venous catheters in the intensive care unit: Association with month of year and week of month. Qual Saf Health Care 2007; 16:252–255

3. Rothschild JM, Landrigan CP, Cronin JW, et al: The Critical Care Safety Study: The incidence and nature of adverse events and serious medical errors in intensive care. Crit Care Med 2005; 33:1694–1700

4. Sanders RC Jr, Giuliano JS Jr, Sullivan JE, et al; National Emergency Airway Registry for Children Investigators and Pediatric Acute Lung Injury and Sepsis Investigators Network: Level of trainee and tracheal intubation outcomes. Pediatrics 2013; 131:e821–e828

5. Nilsson L, Pihl A, Tågsjö M, et al: Adverse events are common on the intensive care unit: Results from a structured record review. Acta Anaesthesiol Scand 2012; 56:959–965

6. Ricard JD, Salomon L, Boyer A, et al: Central or peripheral catheters for initial venous access of ICU patients: A randomized controlled trial. Crit Care Med 2013; 41:2108–2115

7. Greenfield S, Apolone G, McNeil BJ, et al: The importance of co-existent disease in the occurrence of postoperative complications and one-year recovery in patients undergoing total hip replacement. Comorbidity and outcomes after hip replacement. Med Care 1993; 31:141–154

8.VolppKG,GrandeD:Residents’suggestionsforreducingerrorsinteaching hospitals. N Engl J Med 2003; 348:851–855

9. Chapman DM, Cavanaugh SH: Using receiver operating characteris-tic (ROC) analysis to establish the previous experience threshold for critical-procedure competency. Acad Med 1996; 71:S7–S9

10. Hicks CM, Gonzalez R, Morton MT, et al: Procedural experience and comfort level in internal medicine trainees. J Gen Intern Med 2000; 15:716–722

11. Huang GC, Smith CC, Gordon CE, et al: Beyond the comfort zone: Residents assess their comfort performing inpatient medical proce-dures. Am J Med 2006; 119:71.e17–71.e24

12.AngusDC,KelleyMA,SchmitzRJ,etal;CommitteeonManpowerfor Pulmonary and Critical Care Societies (COMPACCS): Caring for the critically ill patient. Current and projected workforce require-ments for care of the critically ill and patients with pulmonary disease: Can we meet the requirements of an aging population? JAMA 2000; 284:2762–2770

13. Mullon JJ, Gajic O, Gali B, et al: The impact of adding 1 month of intensive care unit training in a categorical internal medicine residency program. Crit Care Med 2009; 37:1223–1228

14.KrellK:Criticalcareworkforce.Crit Care Med 2008; 36:1350–135315.KonradC,SchüpferG,WietlisbachM,etal:Learningmanualskillsin

anesthesiology: Is there a recommended number of cases for anes-thetic procedures? Anesth Analg 1998; 86:635–639

16. Nishisaki A, Nguyen J, Colborn S, et al: Evaluation of multidisciplinary simulation training on clinical performance and team behavior dur-ing tracheal intubation procedures in a pediatric intensive care unit. Pediatr Crit Care Med 2011; 12:406–414

17. Bion JF, Barrett H: Development of core competencies for an interna-tional training programme in intensive care medicine. Intensive Care Med 2006; 32:1371–1383

18. Sznajder JI, Zveibil FR, Bitterman H, et al: Central vein catheterization. Failure and complication rates by three percutaneous approaches. Arch Intern Med 1986; 146:259–261

19. Schummer W, Schummer C, Rose N, et al: Mechanical complica-tions and malpositions of central venous cannulations by experienced operators. A prospective study of 1794 catheterizations in critically ill patients. Intensive Care Med 2007; 33:1055–1059

20. Sagarin MJ, Barton ED, Chng YM, et al; National Emergency Airway Registry Investigators: Airway management by US and Canadian emergency medicine residents: A multicenter analysis of more than 6,000 endotracheal intubation attempts. Ann Emerg Med 2005; 46:328–336

21. Jaber S, Amraoui J, Lefrant JY, et al: Clinical practice and risk factors for immediate complications of endotracheal intubation in the inten-sive care unit: A prospective, multiple-center study. Crit Care Med 2006; 34:2355–2361

22.McGeeDC,GouldMK:Preventingcomplicationsofcentralvenouscatheterization. N Engl J Med 2003; 348:1123–1133


Roux et al


23. Ricard-Hibon A, Chollet C, Leroy C, et al: Succinylcholine improves the time of performance of a tracheal intubation in prehospital critical care medicine. Eur J Anaesthesiol 2002; 19:361–367

24. Adnet F, Jouriles NJ, Le Toumelin P, et al: Survey of out-of-hospital emergency intubations in the French prehospital medical system: A multicenter study. Ann Emerg Med 1998; 32:454–460

25. Schwartz DE, Matthay MA, Cohen NH: Death and other complica-tions of emergency airway management in critically ill adults. A pro-spective investigation of 297 tracheal intubations. Anesthesiology 1995; 82:367–376

26. Wayne DB, Barsuk JH, McGaghie WC: Procedural training at a crossroads: Striking a balance between education, patient safety, and quality. J Hosp Med 2007; 2:123–125

27.DurbecO,ViviandX,PotieF,etal:Aprospectiveevaluationoftheuseof femoral venous catheters in critically ill adults. Crit Care Med 1997; 25:1986–1989

28. Merrer J, De Jonghe B, Golliot F, et al; French Catheter Study Group in Intensive Care: Complications of femoral and subclavian venous catheterization in critically ill patients: A randomized controlled trial. JAMA 2001; 286:700–707

29.MouradM,KohlwesJ,MaselliJ,etal;MERNGroup:Supervisingthesupervisors–procedural training and supervision in internal medicine residency. J Gen Intern Med 2010; 25:351–356

30.FragouM,GravvanisA,DimitriouV,etal:Real-timeultrasound-guidedsubclavian vein cannulation versus the landmark method in critical care patients: A prospective randomized study. Crit Care Med 2011; 39:1607–1612

31.Tan TC, Tan KT, Tee JC: An end to “See one, do one and teachone” residency training programme—Impact of the training, educa-tion, surgical accreditation and assessment (TESA) programme on medical care and patients’ safety. Ann Acad Med Singapore 2007; 36:756–759

Acquiring Procedural Skills in ICUs

Documents