correspondence 471

CAN J ANESTH 55: 7 www.cja-jca.org July, 2008

Reply:

We thank Drs. Hemmerling and Carli for their let-ter and for their overall positive view of our paper. We embarked on this study because of the “natural experi-ment” provided by the provincially-mandated consoli-dation of two cardiac surgery units in London into a single heart centre in April 2005. We were pleased to note that the composite morbidity rate for our entire cohort decreased from 16.3% in the year prior to the merger, to 13.0% in the first year after the merger, a dif-ference that was clinically and statistically significant. As Drs. Hemmerling and Carli have noted, most of the improved composite morbidity outcome was observed in patients undergoing isolated coronary artery bypass graft (CABG) surgery. They correctly pointed out that the composite morbidity rate after isolated valve surgery did not change between year 1 and year 2, even though the isolated valve surgery mortality rate decreased from 4.3% to 3.6%. It should be noted, however, that the com-posite morbidity rate in combined CABG-valve patients decreased from 37.1% to 26.4% from year 1 to year 2, a value that bordered on statistical significance and had clinical importance. Drs. Hemmerling and Carli indicated that it was “somewhat uncertain as to which factor(s) were predom-inantly associated with each change in outcome”. Given the fact that this study reported on a “natural experi-ment”, as opposed to a randomized controlled trial, it was very difficult to establish the precise causality among observed outcomes. Nonetheless, on multivariable model-ing the pre-merger year was shown to be an independent predictor of death or major complications, with an odds ratio and P value that were similar to other major risk factors such as a creatinine > 120 umol·L–1 and diabetes. The possibility of bias is inherent in any prospective cohort study, especially in those with a before/after study design. This issue was reviewed extensively in the Discussion of our paper, which also highlighted the specific factors that we believe could have led to the improved outcomes that we demonstrated in the second year of the study. Drs. Hemmerling and Carli’s letter also mentioned the “lack of standardization of the intraoperative anes-thetic technique and surgical management” and the fact that our median time to tracheal extubation was still 8.0 hr in the second year of the study. We hereby con-firm that our intraoperative anesthetic techniques and surgical procedures were as well standardized as those at any other academic health sciences centre, and that our experience reflects the “real world” conduct of contem-porary cardiac anesthesiology and surgery practice. A recent systematic review of fast-track cardiac anesthesia encompassed in its operational definition the intention

to promote early (< 10 hr) tracheal extubation.1 We are pleased to report that our median postoperative ventila-tion hours have continued to decrease successively from 8.0 hr, in the second year of our study to under 7.0 hr presently. As Drs. Hemmerling and Carli correctly point out, comprehensive clinical pathways covering not only ventilator weaning, but also a host of postoperative care elements, are important as cardiac surgery units strive to improve outcomes and decrease resource utiliza-tion.

Richard J. Novick mdmscfrcsc

Stephanie A. Fox bscrrcp

Larry W. Stitt msc Ron Butler mdmscfrcpc

Mary Kroh bscn

Christina Hurlock-Chorostecki mscnacnpcncc(c)Chris Harris rrt

Davy C.H. Cheng mdmscfrcpc

London Health Sciences Center, University Hospital, London, CanadaE-mail: richard.novick@lhsc.on.ca

Reference 1 Myles PS, Daly DJ, Djaiani G, Lee A, Cheng DC. A

systematic review of the safety and effectiveness of fast-track cardiac anesthesia. Anesthesiology 2003; 99: 982–7.

Skin infection and necrosis after subcutaneous infiltration of propofol in the intensive care unit

To the Editor:Microorganisms grow rapidly in propofol, and extrin-sic contamination of propofol is thought to be a source of postoperative sepsis and wound infection.1 We describe a case in which propofol is believed to have caused skin necrosis. A 70-yr-old male underwent coronary artery bypass grafting and repair of the aortic and mitral valves. His postoperative course was complicated by acute renal failure, encephalopathy, Pseudomonas aeruginosa pneumonia, methicillin resistant staphylococcus epi-dermidis (MRSE) bacteremia, and respiratory failure that required prolonged mechanical ventilation. He was sedated with a continuous intravenous infusion of propofol via a 14-G catheter inserted on the dorsum of his right hand. The propofol infusion was handled according to the manufacturer’s directions. Three days

472 CANADIAN JoURNAL oF ANESTHESIA

CAN J ANESTH 55: 7 www.cja-jca.org July, 2008

after the infusion was started, extravasation of propo-fol was noted under the patient’s skin. The catheter was removed and the site dressed. Two days later, the site became infected, then necrotic and ulcer-ated (Figure). A culture of the ulcerated tissue grew Pseudomonas aeruginosa and MRSE. At the time, the patient’s antibiotic therapy consisted of vancomycin and ceftazidime. Ciprofloxacin was added later, to deal with the Pseudomonas, cultured from his skin infection that was resistant to ceftazidime. The necrotic tissue was surgically debrided and the wound healed. Two weeks later, the patient was discharged to a skilled nursing facility. Propofol is an excellent culture medium secondary to its high lipid content. Propofol’s ability to support bacterial growth is clearly stated on the package inserts of the two, commercially available propofol prepara-tions in the United States. Contamination of propo-fol with Candida albicans has been found to cause fungemia and endophthalmitis.2 other organisms that have been related to propofol administration include; Staphylococcus epidermis, Staphylococcus aureus, Serra-tia marcescens, and Klebsiella pneumonia.3

Propofol has been found to cause immuno-modu-lation, weakening natural body defense mechanisms and, hence, increasing the risk of infection. In the intensive care unit, where propofol is often used for an extended period, it may exacerbate immune sup-pression, as seen in critically ill post surgical patients.4 The manufacturers of the two propofol preparations recommend specific precautions for preventing con-tamination of the product. These recommendations include disinfecting the vial’s rubber stopper and minimizing intravenous line manipulations. Adminis-

tration must be completed within 12 hr after the vial has been spiked, and the administration intravenous lines must be changed after six hours. If the propofol emulsion is transferred to a syringe or other container prior to administration, the drug should be discarded. Lorenz et al.5 showed that using such criteria did not completely eliminate contamination, and it was not superior to the usual routine handling of propofol in the operating rooms. The presence of the same organism, in cultures obtained from this patient’s sputum and skin lesion, raises the possibility of hematogenous spread that seeded in the dorsum of his right hand and likely resulted from the presence of this extravasated pro-pofol, especially when the patient’s blood cultures showed the same bacterial growth (MRSE). The other possibility could have been contamination by contact of the skin with infected sputum. It has been proposed to add lidocaine in order to lower this risk of propofol associated infection. Lidocaine possesses bacteriostatic activity against esch-erichia coli and confers its bacteriostatic activity to the mixture. This may decrease the risk of infection from contaminated propofol.1 In contrast, the use of propo-fol alone strongly supports the growth of E. coli and P. aeruginosa. We believe that subcutaneous extravasation of pro-pofol is a risk factor for skin infection, necrosis, and ulceration. Clinicians should closely monitor propo-fol intravenous infusion sites, and, if extravasation is noted, measures should be taken to prevent contami-nation of the site.

Basem B. Abdelmalak md C. Allen Bashour md Jean- Pierre Yared md Cleveland Clinic, Cleveland, USAE-mail: abdelmb@ccf.orgAccepted for publication January 10, 2008.

References 1 Sakuragi T, Yanagisawa K, Shirai Y, Dan K. Growth

of Escherichia coli in propofol, lidocaine and mixtures of propofol and lidocaine. Acta Anaesthesiol Scand 1999; 43: 476–9.

2 McNeil MM, Lasker BA, Lott TJ, Jarvis WR. Postsur-gical Candida albicans infections associated with an extrinsically contaminated intravenous anesthetic agent. J Clin Microbiol 1999; 37: 1398–403.

3 Veber B, Gachot B, Bedos JP, Wolff M. Severe sepsis after intravenous injection of contaminated propofol. Anesthesiology 1994; 80: 712–3.

4 Stevenson GW, Hall SC, Rudnick S, Seleny FL,

FIGURE Dorsum of the patient’s right hand showing the extent and the depth of the skin infected necrosis, and ulceration.

correspondence 473

CAN J ANESTH 55: 7 www.cja-jca.org July, 2008

Stevenson HC. The effect of anesthetic agents on the human immune response. Anesthesiology 1990; 75: 542–52.

5 Lorenz IH, Kolbitsch C, Lass-Florl C, et al. Routine handling of propofol prevents contamination as effec-tively as does strict adherence to the manufacturer‘s recommendations. Can J Anesth 2002; 49: 347–52.

Use of the Endoflex® endotracheal tube as a stylet-free alternative in Glidescope intubations

To the Editor:The Glidescope® Videolaryngoscope (GVL; Verathon Inc., WA, USA) has been widely used for intubation in recent years. Incorporating a high-resolution cam-era into its patented, 60° angulated blade, it allows for good visualisation of the glottis, even in patients with more difficult airways. Due to the unique angulation of the blade, a dedicated manufacturer provided the stylet (Glidescope® Rigid Stylet, Verathon Medical, Bothell, WA, USA) utilized to facilitate intubation.A However, there have been reports highlighting the complications of utilizing such rigid stylets in Gli-descope intubations. These difficulties include soft palate, pharyngeal, and tonsillar lacerations and per-forations.1–3 The Endoflex® endotracheal tube (ETT) (Merlyn Associates, Tustin CA, USA) is a single lumen, disposable ETT with an in-built flexing mechanism and a friction lock that allows for easy articulation of its distal tip over a range of angles, without the need for a separate stylet.B We found that Endoflex® ETT’s distal articulation closely followed the 60° angle of the GVL and the Glidescope Rigid Stylet® (Figure). Consequently, we explored its use, in our institution, as an alternative to the Rigid Stylet in Glidescope intubations. our intubation technique remains fairly similar to the traditional means of intubation with the GVL. First, the Glidescope blade is inserted in the midline, with direct visualization of the passage of the blade, and the laryngeal view is optimized before the Endo-

A Glidescope® Video Laryngoscopes, Verathon Inc, WA, USA. Available from http://www.verathon.com/glidescope_prod-ucts.htm (accessed April 2008).

B EndoFlex® Endotracheal Tube, Merlyn Medical, Tustin CA, USA. Available from http://www.merlynmedical.com/ endoflex.php (accessed April 2008).

flex® ETT is introduced into the oral cavity, also in the midline. Next, by applying traction on the fric-tion lock mechanism, the distal articulating tip of the Endoflex® is flexed, and the ETT is advanced under direct vision until it comes into the field of view of the Glidescope. In order to minimize risks of airway trau-ma, visualization of the ETT tip, as it passes through the oropharynx, is important. The Endoflex®, with its flexible, articulating tip (allowing for changes in angulation if needed to facilitate intubation), is then directed to the glottic opening. once the tip of the Endoflex® has been passed beyond the vocal cords, the friction lock is released, thus allowing the Endo-flex® to passively return to its normal curvature aiding passage into the trachea. The tube is then advanced under vision to complete intubation before the Glide-scope is removed. The Endoflex® ETT appears to be a promising alternative to the Rigid Stylet in Glidescope intuba-tions, with its main advantage being the possible reduction or elimination of airway injuries associ-ated with the use of rigid stylets. The flexible distal articulating tip, which flexes over a range of angles, also appears well suited to Glidescope® intubations. Though significant, airway trauma attributed to use of the Glidescope remains rare in our centre and con-sists, for the most part, of case reports in published literature. Perhaps the Endoflex® ETT will provide a welcome alternative in the airway arsenal of the anes-thesiologist.

Figure 1 The Glidescope® Videolaryngoscope (G), fully articulated Endoflex®

endotracheal tube (E) and the Glidescope® Rigid Stylet (S).

G

E

S

FIGURE The Glidescope® Videolaryngoscope (G), fully articulated Endoflex® endotracheal tube (E) and the Glidescope® Rigid Stylet (S).