Fiberoptic Bronchoscopy in Ventilated Patients

Fiberoptic Bronchoscopy in Ventilated Patients· Evaluation of Cardiopulmonary Risk under Midazolam Sedation Jean-Louis Trouillet, A/.D.; i\1arguerite Guiguet, ~\I.S.; C[aul!e Gibert, ~\1.D.; ]ean- Yves Fagon, ~\I.D.; Didier Dreyfuss, ll.D.; Fran~oise Blanchet, Phann.D.; and Jean Chastre, A/.D.

One hundred seven acutely ill ventilated patients were prospectively studied to ascertain the severity and frequency of alterations in gas exchange and hemodynamic parameters during brief bronchoscopy. Sedation was performed using midazolam (0.1 mglkg IV) without topical anesthesia. An average decline in PaD! of 26 percent was observed at the end of the procedure, compared to the baseline value, and this was associated with a mild increase in PaCO! in spite of the use of a special adapter. Alterations in mean systolic blood pressure appeared to be modest, consisting of a 10 percent decrease from the control level, related to sedation, and a 10 percent rise from baseline during the procedure, associated with a concomitant mild tachycardia. At that time, central hemodynamic measure-

I cfu =colony-forming units I

The approach to the diagnosis of diseases of the chest increasingly involves the use of flexible fiberoptic bronchoscopy. The risk inherent in such an examination appears slight in alert, spontaneously breathing patients, 1-5 althou~h the associated occurrence of cardiac arrhythnlias, hypoxenlia, or laryngospasm is not unusual.6-15 In intubated patients in the leU, the flexible fiberoptic bronchoscope has gained restricted use for the diagnosis and treatment ofaif\\ray problems that occur during mechanical ventilation. The generally accepted indications for bronchoscopy in noninlnlunocompronlised ventilated patients are removal of retained secretions, resolution of atelectasis, and evaluation of hemoptysis. ".Ill Patients with endotracheal tubes are at a particularly high risk of nosocomial pneumonia, and the clinical distinction benveen bacterial colonization and true infection is often difficult. One of the most important aids to the definitive diagnosis of bacterial infection, and, specifically, of bacterial pnelunonia, lies in the proper collection of specilnens. Theref()re, recent

efforts have been Inade to salnple the lo\\rer respiratory tract \vithout contalnination. For this purpose a catheter brush that is protected by a telescoping double sheath \\rith a polyethylene glycol plug and passed throu~h the sllction channel of a flexible fiberoptic bronchoscope has been developed by Wimberley and associates. Ii" This brush has been docliinented to have reasonable precision in diagnosing nosocomial pneumonia \\rhen associated \\rith quantitative culture techniques. lk - 2o Foll<)\ving these observations, the use of brief fiberoptic bronchoscopy has increased dralnatically in leus; ho\vever, the nlorbidity associated \"ith this procedure has been Ininitnally investigated in large cohorts of critically ill patients. 1(,.21 Thus, the purpose of this prospective study \vas to deternline the alterations in blood gas levels and henlodynanlic status induced by brief fiberoptic bronchoscopy in 107 critically ill patients undergoing mechanical ventilation. It \vas also our aiIn to establish, if possible, the main risk factors for the developlnent of hypoxernia and heillodynanlic disturbances.

ments performed in a subset of 31 patients showed a significant increase in cardiac output associated with higher pulmonary wedge pressure. Fourteen patients developed hypoxemia of less than 60 mm Hg on Flo! adjusted to 0.8. Of the ten risk factors univariately associated with hypoxemia, only the presence of ARDS (p
*Fronl

the Service de Reanilnation ~Iedicale, and the Serviee d'Explorations Fonetionnelles, Ilt)pital Bichat~ and the Unite de Recherches Biomathematiques et Biostatistiques, INSER~'I U 263, Paris, Fran<,'e. Supported in part hy a ~rant froln the Faculte Xavier Bichat. Manuscript received Au~ust 9~ revision accepted Novelnher 6.

Reprint rei/ul'sts: !Jl: Trouil/l't, Sl'rt'icl' ell' Rellui11111tion J\ll'elicllll' , Hopital Bichllt, 46 Rul' Ilea'; Iluclwrd, 7.5018 fluis, FrlllU'1'

~1:\TERIALS

AND ~'IETII(»))S

POI1Ulation

Patients eligible fi)r participation were those older than 15 years of age who were hospitalized in the leu of Bichat Ilospital during a 14-nlonth period and who were being ventilated at the tilne of suspicion of bacterial pneulnonia. The dia~nosis of pneulnonia was CHEST / 97 / 4 / APRIL. 1990

927

(.'onsidered if any three of the followin~ clinical variables \llere present: ne\\' infiltrates on chest x-ray films; fever (temperature >38.3°C [100.9°F)); leukocytosis (> 10,OOO/cu mm) or leukopenia «5,OOO/cu mm); and punllent tracheal aspirations. Criteria for exclusion of patients were (1) severe hypoxemia (PaO z <60 mnl H~ on Floz of 0.8), (2) systolic blood pressure less than 85 Inm H~, (3) definite acidemia (pH <7.20), or (4) narrow endotracheal prosth(~sis (internal diameter <7 mm). Three patients were excluded ac(.'()rdin~ to the prolc)(.'ol and t,,'o others because of technical hlood ~as samplin~ (discussed suhsequently). prohlerns durin~ Inforrned consent \\'as ohtained from each subject or from the nearest rt:~lative. A total of 107 consecutiv{~ ventilated patients \\'ere induded. Initial El'lllul,tion Patients were evaluated befi)re the procedure usin~ a special dlart c()ntainin~ the fi)lIo\\rjn~ items: a~e; sex; indication fin v{~ntilation support divided into six classes;22 presen(.'e or absence of dlronic pulrnonary disease; presence or ahsence of underlyin~ cardia(' disease (sylnptornatic or re(-ently operated); on~oin~ ARDS; s('ore (APACIIE);2.1 Glas~()w (.'()nla s('ore; requireacute physiolo~k (epinephrine; dopamine; dohutamine); rnent fi)r inotropic.' a~ents prosthesis; duration of mc<:hank-al ventilation size of end()tracllt~al presence of major hemostatic ahnorat tin1{~ of exanlination (d~lYs); Inalities; presenee of Inajor nletaholic disorders; severity of pulmonary infiltrates evaluated hy a radiolowc s(.'()re;IQ use of PEEP (>5 cm H z(»); expiratory rninute volume (Umin); peak inspiratory airway pressure (em H 20); and dynarnic eomplianee (ealeulated as tidal volume divided by peak inspiratory pressure minus end-expiratory pressure). Pn)c(~dun'

All patients were ventilated throu~h a volume-preset ventilator (Servo 900 C; Sielnens-Elelna). The FIo 2 "'as adjusted to 0.8 at 20 minutes hefi>re premedication and was maintained unchan~ed durin~ the period of stndy. Systemic arterial pressure and arterial an intravaseular catheter hlood sHlnples were obtained throu~h inserted into a radial artery as part of routine care of the patient. The p~ltients were sedated \\rith midnzolam (0.1 mglkg IV) five .ninutes before the exmnination, whieh was performed \\,ithout topical anesthesia. A flexihle fiberoptic bronchos(.'()pe (Olympus BF 83R; OD, 6 I1l1n) was introduced through an ind,,'ellin~ endotraeheal tuhe using a special adapter (Boda'i su('tion-safe Y; Sontek ~tedkal Ine). Fiberoptic hronchos('()py was performed by one of two exptarieneed bronchos{'opists and induded only the hacteriolo~ic smnpling using a telesc..'oping ('annula brush catheter (Meditech model R\VF lOnOI9O), fi)llowed hy direet suction throu~h the inner channel of the endoscope. Sarnples of arterial blood were imnlersed in ice and measured \\rithin 30 .ninntes using a blood gas analyzer (fuldi()nltatt~r ABL 30) at 370C. ~Iinute ventilation, airn'a)' pressure, respiratory rate, and dynamic (.'()mplian(.-e "'ere evaluated nsin~ the pnt~urnotacho~raph of the ventilator. \Vhen a triple-lumen thennodilution Swan-Ganz catheter had been inserted, pulmonary artery and capillary wed~e pressures were nleasured throu~h the monitoring systenl (He\\'lett-Paekard 78342 A). In such (.·ases, cardia(' output was measured usin~ a eardiac output (.'<>nlputer (Edwards Labonltories 9520 A). PnJtocol

Arterial blood gas levels, heart rate, blood pressure, and (.'entral heruodynalnic data were nleasured in the following order: (1) prior to preoperative medication; (2) five minutes after IV premedication; (3) at the end of the procedure, with the fiberoptic bronchoscope still in place; (4) 15 minutes after removing the endos(.'()pe; and (5) two honrs after examination. Ventilator settin~s were maintained c..'()nstant throughout the proto<.'ol. Durin~ bn)nehos(.'()py, changes in ventilation were notk-ed as follows: maximal decrease of minute

928

ventilation re('()rded using the expiratory pneumotachograph of the ventilator; and number of incoordinate movements using variations of ventilator pressure cycle during assisted ventilation. For this latter purpose, 1\\'0 estimates "'ere made: (1) the number of cycles partially aborted using a pop-off pressure set at 70 cm H 20; and (2) the number of cycles with pressure trough level lower than - 10 cycles). em H;20 (tri~ered Statistical Analysis Comparisons of quantitative sequential variables were performed using a two-way analysis of variance and a Ne\\'man-Kenls test. Regression analyses "'ere performed with the least-squares method and snbjeeted to step"'ise linear analysis. Univariate analysis of dichutomous risk factors and hypoxemia defined as PaO z less than 60 mm H~ at any time, or hypotension defint~d as systolic blood pressure less than 80 mIn II~ or a fall of nlore than 30 pereent from the initial value, was aecomplished by computin~ odds ratio and 95 percent confidence intervals. Finally, risk faetors and adjusted multiple lowstic re~ression relative odds were ohtained hy usin~ analysis. The ealeulations were perf()rmed ,,'ith BMDP software (U niversity of California, Los An~eles). RESULTS

Clinical characteristics of the 107 patients are sumnlarized in Table 1. The patients had been admitted to the leu for a variety of problems, including postoperative respiratory failure (31 cases; 29 percent), respiratory insufficiency associated with chronic obstructive pulmonary disease (24 cases; 22 percent), acute pulmonary disease (16 cases; 15 percent), neurologic emergencies (14 cases; 13 percent), drug overdose (6 cases; 6 percent), and other miscellaneous disorders (16 cases; 15 percent). Patients were intubated for an average of 4.6 ± 6. 7 days (range, 0 to 45 days). Thirty-eight patients (36 percent) had bacterial pneumonia as defined by quantitative protected specTable 1- Characteristics of 107 Patients Characteristic A~e, yr Sex ratio (~IIF) History of ehronic.' puhnonary disea4ie, per(''ent cardiae Presence of underlyin~ disea~e, percent Presence of ARDS, pereent APACHE score Glas~ow (.'oma s('()re Treatment with inotropic.' a~ents, percent Presence of hydroelectrolytie disturhanees, percent Radiolowc s(.'()re Use of PEE~ percent Minute ventilation, Umint Peak inspiratory airway pressure, cm H 20t Dynamic (.'()mplianee, mVcm H 20t

~.1ean*

2SEM Median

Ran~e

3

64

20-88

1.5 0.8

14 14

5-53 3-15

5.3 23 10.4 29.7

0.4

5

2-10

0.5

1.9

10 30

5-19 14-58

24.6

1.6

22

11.7-50

61 68/39 34 41 13 16.2 12 28

30

*For quantitative variahles. tMeasured at control time. Fiberoptic Bronchoscopy in Ventilated Patients (Trouillat at 8/)

250 C)

225

:I:

t

E 200 E N

175

a-

150

0 co

oj TS Time

Tl

FIGURE 1. Sequential Ineasurelnents of Pa01 (nlean ± 2 SEM) in ventilated patients undergoing flexihle fiheroptic hronchoscopy; FIo:2 was adjusted to n.B at 20 minntes hefi>re endoscopy and was nlaintained unchanged durin~ study. T I • Control tinIe; T;z. five Ininutes after IV premedication; T:,• at end of pro<..'edure. with fiheroptic hront'hoscope still in place; T.,. 15 minutes after removin~ endoscope; and T~. two hours after exmnination. Asterisk indicates significant differences vs any other tilue (p
inlen brush culture using a threshold of 1(}1 cfu/lnl. The mean duration ufthe fiberoptic bronchoscopy was 120 ± 41 seconds. Arterial Blood

GllS

for the 93 others (p
Alonitoring

in Pa02 are sho\vn in Figure 1. The Pa0 2 was not affected by sedation hut had decreased significantly by the end of the procedure, with a mean decrease of 26 percent from the baseline value (p
Baseline PaC},

:39.:3

Use of PEEP-

Chan~es

25.H

Change in PaC()! :3.5.k Radiologic seore 1.6 Peak inspiratory airway pressure at control tiluto. eln tl l ( ) 1.6 1.6 Baseline pH Absence of chronic pulnlonary disease 1.3 Presence of ARDS 1.3

No relation could be established (F
45 Z35

~

~

f~3

40

'" (5

E c

+ ::I:

Tl

35

"7.40

::I:

Q.

"'C

o

1;4

o

7.45

co

TS

7.50

01

~rl-.

- - - - , - - -__- _ - - . . , . - - . . . . . , . - -__- - . . . . - - _ - _ - _

34

36

38 40 Pa C02 (mm Hg)

42

[

FI<;tIRE 2. Changes in PaC()! and [11+1 (Inean ± 2 SEM) durin~ study. Scqut>ntial tinles are the SaJJle as in Figure 1. Value for T"~ was olnitted hecause of sinlilaritv indicate si~ificant diffe~­ to T I .- Asteri.~ks enees vs any other tinle (p
929

creased during bronchoscopy (136 ± 29 mm Hg to 150±45 mm Hg; p250 mm Hg) \vas recorded in three cases. Central hemodynamic data were obtained through a Swan-Ganz catheter in 31 patients (Fig 4). An increase in cardiac output \\'as observed at the end of bronchoscopy, concomitant with a rise in pulmonary wedge pressure, but without significant variations in systemic resistances. In fact, cardiac output did increase in 13 patients without underlying cardiopathy by a mean value of 24 percent, but not in the other 18 patients. Group data from 30 patients demonstrated a significant decrease in Pv02 from 37.5 ± 8.5 mm Hg before the procedure to 33.7 ± 6.7 mm Hg at the end of bronchoscopy (p
110

c:

"e ..,en (0 ~

.D

*

100

..,Q) (0

a:

..,... (0

90

Q)

:I:

0

160

~~ ~

140

E

Jie

Risk Factors

Fourteen patients developed clinical hypoxemia as defined by Pa02 lower than 60 mm Hg on FIo2 adjusted toO.8. Distribution of the variables associated

120

o

T,

TS Time

8

FIGURE 3. Sequentiallueasurements of heart rate and systolic blood pressure (mean ± 2 SEM) in ventilated patients under~oin~ flexible fiberoptic hronchos(.''Opy. Times are the ~lme as in Fi~ure I. Astlnsk.'i indicate p
\vere factors independently related to the degree of relative hypoxemia as determined by using the multiple regression model. As shown in the tabulation, the decrease in Pa02 was related to the severity of pulmonary injury as identified by the level of initial Pa02 on FIo2 adjusted to 0.8 and use of PEE~ Alteration of alveolar ventilation was the other contributing factor. The relative weight of ARDS was obviated by the linked variable, PEE~ The increase in PaC02, which was used as an indicator of alteration of alveolar ventilation, was independently related to the number of pop-off cycles (F = 25.2), the decrease in minute ventilation (F = 10.8), the number of triggered cycles (F = 8.4), changes in Pa0 2 (F = 7.8), and radiologic score (F = 5.1). Henwdynanlic Monitorilll!,

Variations of heart rate and systolic blood pressure are indicated in Figure 3. The heart rate increased slightly at the end of the procedure, with an average of 103 beats per minute, but wide interindividual variations were observed (range, 30 to 170 beats per minute). Major arrhythmias were observed in six patients (tachyarrhythmia in four; bradycardia less than 50 beats per minute in one; multiform ventricular arrhythmia in one). The mean systolic blood pressure fell after sedation by approximately 10 percent, in930

c

7

'E

.... 6

8

5

o -20 en :I:

E E

~

~

* 15

u ~

0 It)

IE (J

.......

1300

VI

.......

c

>

~

ex:

>

(J)

1100 900

OJ

T,

FIGURE 4. Main central hemodynamic data (mean ± 2 SEM) in 31 patients in whom Swan-(;anz catheter had heen inserted. Note that these patients were more severely ill than others (APACHE score, 19.5; nldioloJ.,ric score, 6.6; PaD:! at TI, 166 mm H~; ARDS present in ten; and 18 patients required inotropic agents). CO, Cardiac pulmonary ('apillary wedge pressure; and SVR, output; PC\V~ systemic vascular resistance. Asterisks indicate significant differences vs any other time (p
Table 2-Risk Factors Associated with Hypoxemia by Univariate Analysis in Patients Undergoing Flexible Fiberoptic Bronchoscopy No. of Patients with Hypoxemia (percent)*

No. of Patients at risk

Risk Factor Before procedure A~e <50 yr History of chronic pulmonary disease ARDS APACHE s(.'Ore > 18 Glas~ow coma score <10 Radi()lo~ic score >5 Use of PEEP >5 cm H 2 O Minute ventilation> 10 Umin Pa0 2 (FIoJ = 0.8) <200 mm H~ Heart rate> 100 beats per minute Size of tracheal tube <8 mm Durin~ pnK..'edure Respiratory cycles with elevated inspiratory pressure> 10 No. of triggered cycles >3 Maximal decrease in minute ventilation >60 percent Duration of fiberoptic bronchoscopy >120 s

(30) (3) (57) (14)

Cnlde Odds ratio

95 Percent Cc)nfidenct'

Liluits

p Vailit'

1.6-16.8 0.01-0.8 5.0-74 0.4-3.9 0.1-1.9 2.2-32.8 2.1-22.8 1.4-21 1.2-14.1 1.2-14.1 0.2.'5-5

O.(KJ6 0.03 O.(KK) I 0.77

O.(K)2 0.009 0.02 0.02 0.97

27 36 14 35 28 39 23 48 45 45 92

8 1 8 5 2 11 8 11 10 10 12

(28) (35) (23) (22) (22) (13)

19.3 1.2 0.4 8.5 6.9 5.5 4.1 4.1 1.0

54

12 (22)

7.3

1.5-34.4

0.008

31 58

8 (26) 11 (19)

4.0 3.6

1.2-12.7 0.9-13.7

0.02 n.OR

42

7 (17)

1.6

0.5-4.9

0.39

(7)

5.2 0.1

O.:3.1j O.(XX~

*Hypoxemia (n = 14) was defined as Pa0 2 1ess than 60 mm Hg using Flo 2 of 0.8.

with such hypoxemia is shown in Tables 2 and 3. Using univariate analysis, ten factors could be identified. After stepwise logistic regression analysis, two independent variables were significantly associated with hypoxemia. Eight of the 14 patients with ARDS and 12 of 54 who had frequent pop-off cycles during fiberoptic bronchoscopy had a Pa02 lower than 60 mm Hg. When considering the risk of hypotension, defined as systolic blood pressure less than 80 mm Hg or a decrease of more than 30 percent from the initial value, only two variables were linked using univariate calculations. The APACHE score, with a cutoff of 18 points, was the main predictor (p<0.OO5). Systolic blood pressure lower than 120 mm Hg during the control period was also associated with an increased risk of developing hypotension (p
Risk Factor ARDS Respiratory cycles with elevated inspiratory pressure > 10 durin~ bronchos(.'Opy

Adjusted Odds Ratio

95 Percent Confidence Interval

p Value

15.0

3.7-60

0.001

5.1

1.0-27

0.05

analysis, gave odds ratios (95 percent confidence intervals) of 3.7 (1.1 to 12.2), and 4.3 (1.3 to 14.2), respectivel~ DISCUSSION

Fiberoptic bronchoscopy is generally regarded as safe, based on surveys of endoscopists. Such studies provide important data for alert patients undergoing diagnostic investigation 1-9.14.15 but fail to adequately identify factors that may increase the risk of endoscopy in ventilated patients. As we approach an era \vhere bronchoscopy may supersede bronchial aspiration as a sampling procedure for diagnosing pneumonia in intubated patients, the number of endoscopic procedures will undoubtedly increase, and even slnall rates of complications will become Inagnified. Therefore, recognition of factors that Inay enhance the safety of fiberoptic endoscopy is crucial. This study indicates that a decline in Pa():.! is consistently associated \vith fiberoptic bronchoscopy in ventilated patients and that this decline can he substantial, at least \vhen using Inidazohull as the unique sedative drug. The nlean changt.~ in PaO:.! from baseline during bronchoscopy \vas 26 perct~nt by the end of the procedure; ho\\'ever, large interindividual variations in Pa0 2 changes \\'ere noticed. With regard to our results, the degree of hypoxenlia induced by fiberoptic bronchoscopy \\'as linked to the severity of pulmonary dysfunction and the decrease in alveolar ventilation. This decrease in Illinute CHEST I 97 I 4 I APRIL, 1990

931

ventilation was observed despite the use of a special adapter.25-28 Contrary to the results of Lindholm et al,21 we were unable to find a relation between the size of the tracheal tube and the degree of hypoxemia (F<0.5). Because of the short duration (120 seconds) of the procedure in our study, detailed information concerning the respective implication of each time of examination could not be obtained; however, suctioning throu~h the fiberoptic bronchoscope may be a critical factor. 21,29 This point needs further study using intra-arterial oxygen electrodes,30 transcutaneous oxygen monitoring,31.32 or mixed Sv02 monitoring. 33 Clinical hypoxemia, as defined by Pa02 lo\\'er than 60 mm H~, \\'as more frequent in patients with ARDS and in those \\,ho "fou~ht" the ventilator during the procedure, as shown by multivariate analysis. In our opinion, these results do not indicate that sampling throu~h the fiberoptic bronchoscope should not be used to dia~nose bacterial infections in patients with ARDS, because nosoconlial pneumonia is particularly frequent and severe in this setting. 20 The results simply imply that these patients are at higher risk of definite hypoxemia and should be monitored with particular care. Moreover, incoordinate respiratory movements could be abolished using a better anesthetic protocol. Although different groups have recommended that lidocaine not be instilled through the bronchoscope, 17 the rationale for this approach has recently been challenged. 34 .3.,) Therefore, the use of a high dose of nebulized anesthesia using preservative-free lidocaine might be advocated. An alternative approach, which is our current policy, is the addition of a short-acting neuromuscular blocking agent such as vecuronium. Increasing tidal volume might also be considered,21.36 according to the basal level ofmaximal peak inspiratory pressure. 37 As a group, ventilated patients did not demonstrate major blood pressure variations or tachycardia. The mean decrease in systolic blood pressure belo\\' baseline values was 10 percent, which was related to sedation, and not to the bronchoscopic procedure. In normal humans, midazolam at a dose of 0.15 mglkg IV produces a similar response, associated with an increase in heart rate and a decrease in systemic vascular resistance.:lS In our study, definite hypotension \\'as linked to premedication and to the severity of illness. The APACHE score, \vhich is a classification system representin~ the degree of acute illness, and the relative hypotension before the procedure \vere the more appropriate of the 27 itelTIS included in the multivariate analysis. The use of PEEP or the requirement for inotropic agents \vas not a significant risk factor. During endoscopy, a significant rise in arterial pressure \\'as observed (10 percent above control values), associated with an increase in pulse rate of 5 932

percent. Such observations have been observed in routine 6.beroptic bronchoscopy using an intravenous anesthetic technique of intermittent thiopental (thiopentone) and succinylcholine (suxamethonium), but not in patients receiving either Fentanyl or alfentanil. 39 It is unlikely that using higher doses of midazolam would have prevented such modifications in the ratepressure product, which are consistent with those observed by Boralessa et al40 in intubated patients after a bolus of 0.3 mglkg IV for induction of anesthesia. Therefore, alternative anesthetic protocols should be built up in order to minimize risks ofcardiovascular impairment. Major arrhythmias \\'ere noticed in 5 percent of the patients, which agrees with a number of studies in spontaneously breathing patients,3.10-12.15 but is lower than the 40 percent rate observed by Katz et al. 13 In the latter report, arrhythmias \vere frequently associated with a period of major oxygen desaturation of up to 44 percent. Such hypoxemia was not present in our patients; the mean Sa02 was 97 percent (range, 93 to 100 percent) in the six patients suffering arrhythmias. In conclusion, this study has shown that fiberoptic bronchoscopy is practicable in critically ill intubated patients. In the present series of 107 ventilated patients, no death or cardiac arrest occurred during or within the two hours immediately following the procedure; however, patients in the ICU are at risk of relative hypoxemia during fiberoptic bronchoscopy even when high levels of oxygen are provided to the ventilator and gas leaks around the endoscope are minimized by a special adapter. Patients with ARDS are obviously at higher risk of definite hypoxemia. CarefullTIethodic attention to the anesthetic protocol, ventilator settings, and monitoring of patients during bronchoscopy should permit rapid correction and more frequent prevention of hypoxemia and should further decrease morbidity. \Ve thank Drs. Abdel Akesbi and Patrick Bnln for invaluable eooperation during the bronchoseopic procedures they performed, and Miss M. Kail and Miss C. Bnln for patient preparation of the manuscript. ACKNO'VLEDG~fENTS:

REFERENCES 1 Credle \VF Jr, Smiddy JF, Elliott Re. Cornplications of fiberoptic hronehoscopy. Am Rev Respi r Dis 1974; 109:67-72 2 Sundt PM, Smiddy JF, Cruller B. Deaths and ('omplications associated \\rith fiberoptic hronchos('opy. Chest 1976; 69:747-51 3 Pereira W Jr, Kovnat OM, Snider CL. A prospective cooperative study ofeomplications f()lIowing flexihle fiberoptic hronchos('opy. Chest 1978; 73:813-16 4 Fulkerson \Vj. Fiheroptic hronchos('opy. N Eng) J Med 1984; 311:511-15 5 Nil LBI workshop summaries: summary and recommendations of a workshop on the investigative use of fiberoptic bronchoscopy and bronchoalveolar lavage in asthmatics. Am Rev Respir Dis Fiberoptic Bronchoscopy in Ventilated Patients (Trouillet at 8/)

1985; 132:180-82 6 Kleinholz EJ, Fussell J, McBrayer R. Arterial blood gas studies during fiberoptic bronchoscopy. Am Rev Respir Dis 1973; 108:1014 7 Albertini RE, Harrell JH II, Kurihara N, Moser KM. Arterial hypoxemia induced by fiberoptic bronchoscopy. JAMA 1974; 230: 1666-67 8 Albertini RE, Harrell JH II, Moser KM. Management ofarterial hypoxemia induced by fiberoptic bronchoscopy. Chest 1975; 67:134-36 9 Dubrawsky C, Awe RJ, Jenkins DE. The effect of bronchofiberscopic examination on oxygenation status. Chest 1975; 67:137-40 10 Luck JC, Messeder OH, Rubenstein MJ, Morrissey WL, Engel TR. Arrhythmias from fiberoptic bronchoscopy. Chest 1978; 74:139-43 11 Shrader D L, Lakshminarayan S. The effect of fiberoptic bronchos(,'Opy on cardiac rhythm. Chest 1978; 73:821-24 12 Elguindi AS, Harrison GN, Abdulla AM, Chaudhary BA, Vallner JJ, Kolbeck RC, et ale Cardiac rhythm disturbances durin~ 6beroptic bronchoscopy: a prospective study. J Thorac Cardiovasc Sur~ 1979; 77:557-61 13 Katz AS, Michelson EL, Stawicki J, Holford FO. Cardiac 6heroptic bronchoscopy and arrhythmias: frequency durin~ eorrelation with hypoxemia. Arch Intern Med 1981; 141:603-06 14 Matsushima Y, Jones RL, King EG, Moysa G, Alton JDM. Alterations in pulmonary mechanics and ~as exchan~e durin~ routine 6beroptic bronchoscopy. Chest 1984; 86:184-88 15 Payne CB Jr, Goyal PC, Gupta SC. Effects of transoral and transnasal 6beroptic bronchos(,'opy on oxygenation and eardiac rhythm. Endoscopy 1986; 18:1-3 16 Barrett CR Jr. Flexible 6beroptie bronchoscopy in the critically ill patient: methodology and indications. Chest 1978; 73(suppl):746-49 17 Wimberley N, Faling LJ, Bartlett JG. A 6beroptic bronchoscopy technique to obtain uncontaminated lower respiratory secretions for bacterial culture. Am Rev Respir Dis 1979; 119:337-43 18 Chastre J, Viau F, Brun ~ Pierre J, Dauge MC, Bouchama A, et al. Prospective evaluation of protected specimen bnlsh for the diagnosis of pulmonary infections in ventilated patients. Am Rev Respir Dis 1984; 130:924-29 19 Fagon JY, Chastre J, Hance AJ, Guiguet M, Trouillet JL, Domart Y, et aI. Detection of nosocomial lung infection in ventilated patients: use of a protected specimen brush and quantitative culture techniques in 147 patients. Am Rev Respir Dis 1988; 138:110-16 20 Faling LJ. New advances in diagnosing nosocomial pneumonia in intubated patients: part 1. Am Rev Respir Dis 1988; 137:25355 21 Lindholm CE, Oilman B, Snyder J~ Millen EG, Grenvik A. Cardiorespiratory effects of flexible 6beroptic bronchos(,'Opy in critically ill patients. Chest 1978; 74:362-68 22 Zwillich C\~ Pierson OJ, Creagh CE, Sutton FD, Schatz E, Petty TL. Complications of assisted ventilation. Am J Med 1974;

57:161-70 23 Knaus \VA, Zimmerman JE, Wagner O~ Draper EA. Lawrence DE. APACHE: acute physiology and chronic health evaluation: a physiologically based classification system. Crit Care Med 1981; 9:591-97 24 Brackett NC Jr, Cohen JJ, Schwartz 'VB. Carbon dioxide titration curve of normal man: effect of increasing degrees of acute hypercapnia on acid-base equilibrium. N En~d J Med 1965; 272:6-12 25 Renz LE, Smiddy J\~ Rauscher CR, Kerby GR, Ruth \VE. Bronchoscopy in respiratory failure. JAMA 1972; 219:619 1972; 26 Tahir AH. Bronchoscopy in respiratory failure. JA~fA 220:725 27 Shinnick J~ Johnston RF, Oslick T. Bronchosc..'Opy during mechanical ventilation using the fiherscope. Chest 1974; 65:613-15 28 Bodai BI, Blaisdell FW Discontinuance of positive end-expiratory pressure during suctioning: effect on oxygenation in patients with adult respiratory distress syndrome. Surg Fonlm 1981; 32:311-12 29 Lundgren R, Haggmark S, Reiz S. lIemodynanli<.- eflt-cts of flexible 6beroptic bronchosc'opy performed under topical anesthesia. Chest 1982; 82:295-99 30 Arai T, Hatano Y, Komatsu K, Takada T. Miyake C, lIarioka T. et al. Real-time analysis of the chan~e in arterial oxygeu tt"nsiou durin~ endotracheal suction with a fiheroptic hronchos('op{". Crit Care Moo 1985; 13:855-58 31 Berman L, Jones NL. Fiheroptic hronehos('opy: rnouitoring oxygenation with ear oximetry. Anl Hev Respir IJis HJ76; 114:651 32 Gho\\'s MB, Rosen MJ, Chuang MT, Sac..'ks liS, Teirstein AS. Transcutaneous oxygen monitoring during 6heroptic hrouchosc..'Opy. Chest 1986; 89:543-44 33 Walsh JM, Vanderwarf C, Hoscheit D, Fahey PJ. lJ nsusfXacted hemodynamic alterations during endotracheal snctioning. Chest 1989; 95:162-65 34 Kirkpatrick MB, Sanders R~ Bass JB Jr. Physiologic effects and senlm lidocaIne concentrations after inhalation oflidocairlta from a compressed gas-powered jet nebulizer. Am Rev Respir IJis 1987; 136:447-49 35 Berger R, MeConnell J~ Phillips B, Overman TL. Safety and efficacy of using high-dose topical and nebulized anesthesia to obtain endobronchial cultures. Chest 1989; 95:299-303 36 Burnham SC, Martin WE, Cheney F\\Z The effects of various tidal volumes on gas exchange in pulmonary ederna. Anesthesiolo~ 1972; 37:27-31 37 Petersen G~ Baier H. Incidence of pulmonary barotraurua in a medical ICU. Crit Care Med 1983; 11:67-9 38 Reves JG, Fra~en RJ, Vinik HR, Greenblatt OJ. ~tidazolanl: pharmacology and uses. Anesthesiology 1985; 62:310-24 39 Wark KJ, Lyons J, Feneck RO. The haenl
CHEST I 97 I 4 I APRIL, 1990

933