ANESTH ANALG 1997;85:1176-82

LETTERS

TO THE EDITOR

1177

those issues will also not be resolved without appropriately designed, well conducted, randomized trials evaluating relevant outcomes in appropriate patient populations. Kenneth J. Tuman, MD Department of Anesthesiology Rush Medical College Chicago, IL 60612 Michael F. Roizen,
MD and Critical Care Department of Anesthesia University of Chicago Chicago, IL 60612

References
1. Connors AF, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the in&d care of critically ill patients. JAMA 1996;276:889-97. 2. Connors AF, Hare11 FE, Knaus WA, Wagner D. Effectiveness of right heart catheterization: time for a randomized trial [letter] JAMA 1997;277zl14. 3. Bone RC. Effectiveness of right heart catheterization: time for a randomized trial. [letter] JAMA 1997;277:114. 4. D&n jE, Bone RC. Is it time to pull the pulmonary artery catheter? JAMA 1996:276:916-a. 5. The American Society of Anesthesiologists Task Force on Pulmonary Artery Catheterization. Practice guidelines for pulmonary artery catheterization. Anesthesiology
1993:78:380-4.

To this end, guidelines for endotracheal intubation abound. For the patient with a difficult airway, there is an algorithm describing how best to approach ventilation and intubation (1). Miller et al. (2) reviewed the physiology of extubation and described specific techniques for extubation. In an attempt to provide guidelines for our residents regarding whether to extubate a patients trachea in the operating room, we devised an algorithm for an approach to extubation (Figure 1). The algorithm is not exhaustive in that it does not include all possible patient, surgical, or anesthetic conditions. Certainly, as with all other algorithms, the clinicians judgment is of utmost importance in the decision of whether to proceed with extubation. It may, however, serve as a useful educational tool for residents in the fine art of removing the endotracheal tube that was so adeptly inserted at the beginning of an anesthetic. Cynthia A. Lien, MD Howard Koff, MD Vinod Malhotra, MD Farida Gadalla, MD
Department of Anesthesiology The New York Hospital-Cornell New York, NY 10021 Medical Center

References

Emergence and Extubation: A Systematic Approach

To the Editor: Much effort in the education of anesthesia residents is devoted to learning how and when to intubate and extubate patients tracheae.

1. Benumof JL. Management 1991;75:1087-110. 2. Miller KA, Harkin CP, Bailey 1995;80:149-72.

of

the

difficult

adult tracheal

airway. extubation.

Anesthesiology An&h Analg

FL. Postoperative

Anesthetic
To the Editor:

Agents and Platelet Aggregation

Emergence

and Extubation:

A Systematic

Approach

Can this patient be etibated while deeply anesthetized?

.Naresidualne=omuscularblock . Easy mask ventilation . Easily mtubated . Not at increased risk for regurgitationlasp~~atmn
. Normothemdc

Difficult mask vent&ion Difficult intubation l Residual neunxnuscular block Present . Full stomach
l l

- pregIw.nt -obese - recent ingestion of food - diabetic -has ascites /

Can the patient he extubated immediately following surgery and emergence from general anesthesia?

6 . Awake

& . Hypoxic (02 saturation < 90 mmHg)

. Following canmads . Breathing spontanwusly -well oxygenated -not excessively hypercarbic

@CO2 s 50 lnnag)

. Excessively hypercarbic (PaCOI > 50 mmHg)

. Hypothnmic
l

(<34T)

* Residual neuromuscular block present Patient may be unable to pro&t his 01 her own ainvay . Airway swelling -long surgery in Trendelenberg position - ailway surgery
-patient received excessive intravenous fluid volume . ImpaIlment of coughigag reflex - brainstem surgery - intraop cerebral ischmic events m Vocal cord paralysis - Inadequate strength

We read the report by De La Cruz et al. (1) with interest. We have previously investigated the effects of propofol, halothane, and isoflurane on platelet aggregation in vim using electrical impedance aggregometry (2). Platelet aggregation evaluated in whole blood without centrifugation by impedance aggregometry is measured in a more physiological milieu than the mere preparation of plateletrich plasma and additional centrifugation, which results in evaporation of volatile anesthetics as in optical aggregometry. However, our in vim results do not correlate with the in vitro results of De La Cruz et al. (l), and we did not detect any inhibitory effects of propofol, halothane, and isoflurane on platelet aggregation. On the other hand, previous investigations by many authors concerning volatile anesthetic effects on platelet aggregation have all used optical aggregometry (3,4) because impedance aggregometry is a relatively new method. The diversity of these results can also be explained by the fact that pharmacokinetics in vim is quite different than the test tube, with wide variations in drug concentrations (5). We can say that high levels of anesthetics are prone to affect thrombocyte function, but this experimental environment does not take place in the human body. For this reason, further controlled studies of the two methods, using in vim and in vitro conditions simultaneously, are still required to make a definite conclusion. Yesim Ates, MD Yiiksel Kecik, MD Sema Yavuzer, MD
Departments of Anesthesiology b Reanimation Ankara University Medical Faculty Ankara, Turkey and Physiology

References
1. De La Cruz Jl, Carmona JA, Paez MV, et al. lropofol inhibits in vitro platelet aggregation in human whole blood. An&h Analg 1997;84:919-21. 2. Ateg Y, Kqik Y, Yavuzer S. The effects of halothane, isoflurane and propofol anesthesia on platelet aggregation in viva [abstract]. llth World Congress of Anaesthesiologists April 14-20, 1996, Sydney, Australia. 3. Stengert KB, Sellick CL, Lazerson J. Halothane induced platelet dysfunction. An&h Analg 1982;61:217-23. 4. Fuss BG, Meadows JC, Bruni CY, Quereshi GD. The in vitro and in viva effects of isoflurane and nitrous oxide on platelet aggregation. An&h Analg 1986;651170-4. 5. Fisher DM. (Almost) Everything you learned about pharmacokinetics was (somewhat) wrong! An&h Analg 1996;83:901-3.

Excessively long surgical procedures * Airway may be difficult to reestablish * Unexplamcd hemodynamic instabibty
l

I The patient requires continued intubatian

and mechanical ventilation Figure

1. An approach

to extubation.

1178

LETTERS

TO THE

EDITOR

ANESTH ANALG 1997;85:1176-82

In Response: There are obviously controversies about the possible effect of several anesthetics on platelet aggregation, perhaps due to the different technical approaches and interpretations of the results from study to study. Our investigation was in vitro. In no part of our paper do we extrapolate directly to in vivo conditions. Our results only describe a different behavior of propofol in vitro between two types of samples: isolated platelets and whole blood. In whole blood, erythrocytes and/or leukocytes play an important role in platelet function and in its inhibition by drugs. The in vitro effect of propofol is influenced by the nitric oxide-cGMP pathway interaction with platelet-leukocyte (1). For that reason, we think that whole blood aggregometry could demonstrate better than isolated platelets the inhibitory effect of propofol. Even more, we obtained recent in vivo results in whole blood aggregometry in patients who received a bolus of propofol (1). In these patients, platelet function in isolated platelets was not modified statistically, but using whole blood aggregometry, we observed an inhibitory effect 5 min after the propofol bolus (50% effective dose for collagen 0.5 pg/mL before bolus, 1.9 pg/mL after bolus); moreover, plasma nitrites (as indicators of nitric oxide production) were increased after the propofol bolus. In conclusion, our study is the first step of an investigation that requires other consecutive steps to obtain a definitive conclusion. J.P. De La Cruz J.A. Carmona M.V. Paez E. Blanc0 F. Sanchez De La Cuesta Department of Pharmacology and Therapeutics School ofMedxine University of Mdlagu 29071 Mdlaga, Spain References
1. Carmona JA, De La Cmz JP, Paw. MV, et al. Propofol Inhibits platelet aggregation m human whole blood influence of NO-cGMP pathway [abstract]. XXIII Congress of the Spanish Society of Anesthesia. Zaragoza, Spain, June 1997.

a patients arm. The patient sustained blunt trauma to the abdomen, presented in shock to the emergency center, was resuscitated, and was hemodynamically stable on transfer to the operating room. In the OR, the patient was placed on the table with his arms abducted at 90 on padded armboards. Vital signs, including blood pressure, were normal. Before induction, we repositioned the patients arms to his sides, using elbow padding and the draw sheet. The NIABP monitor (Hewlett Packard, Andover, MA) failed to measure the blood pressure and did not recycle automatically after this failure. The unwavering signal from the Spo, monitor, applied to the arm on which we had placed the blood pressure cuff, alerted us to this failure. Trouble-shooting included examining the tubing for kinks, changing the NIABP electrical module, palpating the NIABP bladder during manual cycling of the device, and, finally, changing the blood pressure cuff. After removing the cuff, we noted a small hole at the junction of the nipple fitting and the bladder in the NIABP cuff (DURA-CUFTM; Johnson & Johnson, Arlington, TX; Fig. 1). The leak was not present until the patients arm was repositioned and shear forces partially detached the nipple from the cuff. The leak prevented an automatic determination of the blood pressure. The alarm for such a failure was not obvious in the noisy operating room. Had we placed the pulse oximeter on the side away from the NIABP, as is our usual practice, we would not have had a warning of the failure until after the induction of anesthesia. We have asked the manufacturer to modify the alarm for this type of malfunction. Bernard P. Gallacher, AB, MD.CM, Lewis A. Coveler, BA, MD, DABA Department of Anesthesiology Baylor College of Medicine Houston, TX 77030
CCFP, FRCPC, DABA

Full Disclosure
To the Editor:

in Study Design Is Essential

Failure of a Noninvasive Automated Blood Pressure Monitor to Detect the Blood Pressure After Repositioning of the Patients Arm
To the Editor: We describe the failure of a noninvasive automated (NIABP) monitor to detect the blood pressure after blood pressure we repositioned

Figure bladder

1. Note the small hole in the NIABP cuff.

at the junction

of the nipple

fitting

and

the

Recent correspondence regarding the paper by Shore-Lesserson et al. (1) raised several issues that merit further comment. Although not mentioned in the Methods section (11, a one-tailed test was used to both estimate the sample size and analyze the data (2). One-tailed testing may be justified under specific circumstances; however, at the very least, one-tailed testing should be disclosed wherever it is used; for example, when the hypothesis of the study is defined, in the assumptions for the sample size estimation, and in the post hoc testing criteria. None of these occurred in this paper (1). In addition, as sample size estimation becomes more commonplace in clinical research, it is important that all of the assumptions used are disclosed to the readers. In addition to specifying the LYand 0 values, the effect size must be clearly defined (3). In their paper (11, the authors used a 33% decrease in transfusions between tranexamic acid and placebo as the minimal desired effect size. However, this criterion may be interpreted in at least two ways: 1) an arithmetic difference of 33% in the transfusion rate between the two groups (i.e., from 66% to 33%) or a fractional difference of 33% (i.e., a difference of 66% X 33%). Because sample size increases exponentially as the size of the larger proportion decreases (for a fixed percent decrement) (31, the rate of transfusion that they used could only be determined by back-calculating the rate based on a sample size of 20 patients per group. The authors should have reported the exact incidence that they used, as well as the source of their data. Failure to disclose these data limits our ability to judge whether their assumptions are clinically reasonable, makes it difficult to verify their sample size calculation, and limits the educational opportunity for trainees. Both the readers and the reviewers would benefit enormously if the authors clearly presented all of the assumptions in their study design. Finally, what about one-tailed testing? Although it is limited by its inability to interpret a finding opposite to that defined in the hypothesis, it has the advantage that it decreases the sample size by 20%-25% compared with a two-tailed test. Thus, fewer patients would be enrolled, randomized, treated, and possibly harmed by