| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Pulmonary and Critical Care, Bridgeport Hospital and Yale University School of Medicine, Bridgeport, CT.
Correspondence to: Constantine A. Manthous, MD, FCCP, Bridgeport Hospital, West Tower 6, 267 Grant St, Bridgeport, CT 06610; e-mail: pcmant{at}bpthosp.org
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: We conducted a prospective study of 91 adult patients treated in medical-cardiac ICUs who were recovering from respiratory failure, had successfully completed an SBT, and were about to be extubated. A number of demographic and physiologic parameters were recorded with the patient receiving full ventilatory support and during the SBT, just prior to extubation. Cough strength on command was measured with a semiobjective scale of 0 to 5, and the magnitude of endotracheal secretions was measured as none, mild, moderate, or abundant by a single observer. In addition, patients were asked to cough onto a white card held 1 to 2 cm from the endotracheal tube; if secretions were propelled onto the card, it was termed a positive white card test (WCT) result. All patients were then extubated from T-piece or continuous positive airway pressure breathing trials. If 72 h elapsed and patients did not require reintubation, they were defined as successfully extubated.
Results: Ninety-one
patients with a mean (± SE) age of 65.2 ± 1.6 years, ICU admission
APACHE (acute physiology and chronic health evaluation) II score of
17.7 ± 0.7, and duration of mechanical ventilation of 5.0 ± 0.5
days were studied over 100 extubations. Sixteen patients could not be
extubated, and 2 patients underwent two unsuccessful extubation
attempts, for a total of 18 unsuccessful extubations. Age, severity of
illness, duration of mechanical ventilation, oxygenation, rapid shallow
breathing index, and vital signs during SBTs did not differ between
patients with successful extubations vs patients with unsuccessful
extubations. The WCT result was highly correlated with cough strength.
Patients with weak (grade 0 to 2) coughs were four times as likely to
have unsuccessful extubations, compared to those with
moderate-to-strong (grade 3 to 5) coughs (risk ratio [RR], 4.0; 95%
confidence interval [CI],1.8 to 8.9). Patients with
moderate-to-abundant secretions were more than eight times as times as
likely to have unsuccessful extubations as those with no or mild
secretions (RR, 8.7; 95% CI, 2.1 to 35.7). Patients with negative WCT
results were three times as likely to have unsuccessful extubations as
those with positive WCT results (RR, 3.0; 95% CI, 1.3 to 6.7). Poor
cough strength and endotracheal secretions were synergistic in
predicting extubation failure (Rothman synergy index, 3.7; RR, 31.9;
95% CI, 4.5 to 225.3). Patients with
PaO2/fraction of inspired oxygen (P:F) ratios
of 120 to 200 (receiving mechanical ventilation) were not less likely
to be successfully extubated than those with P:F ratios of > 200, but
those with hemoglobin levels 
10 g/dL were more than five times as
likely to have unsuccessful extubations as those with hemoglobin levels
> 10 g/dL.
Conclusions: After patients recovering from respiratory failure have successfully completed an SBT, factors affecting airway competence, such as cough strength and amount of endotracheal secretions, may be important predictors of extubation outcomes. Also, a majority (89%) of medically ill patients with P:F ratios of 120 to 200 (four of five patients with P:F ratios from 120 to 150), values sometimes used to preclude weaning, were extubated successfully.
Key Words: critical care • critical illness • extubation • mechanical ventilation • weaning
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
7 cm
H2O for 0.5 to 2 h. Bedside clinicians
prematurely terminate SBTs for the following reasons: severe distress
despite attempts of bedside personnel to attenuate anxiety
nonpharmacologically, increment of heart rate > 20 beats/min or
systolic BP > 20 mm Hg, respiratory rate > 35 breaths/min,
VT < 300 mL, or pulse oximetry desaturation to < 90%
while inspiring 50% oxygen. Patients who successfully complete an SBT
are further assessed with an arterial blood gas analysis; if the result
is favorable, the patient is considered for a trial of endotracheal
extubation. In our ICU, some patients successfully complete the SBT and
are not extubated if the clinicians feel they still require an
artificial airway. Patients were excluded from this study if they were
being extubated to comfort care (withdrawal of life-sustaining
therapies, no reintubation).
Measurements
The following data were gathered: age, ICU admission APACHE
(acute physiology and chronic health evaluation) II score, duration of
endotracheal intubation/mechanical ventilation, arterial blood gas
levels on full ventilatory support prior to the SBT, rapid shallow
breathing index (RSBI; respiratory rate divided by VT), SBT
modality, respiratory rate, heart rate, and mean arterial pressure
after roughly 30 min of the SBT, arterial blood gas levels during the
SBT, and presence of congestive heart failure or gross
encephalopathy.
A single observer (M.K.) worked with bedside nurses and respiratory therapists to assess voluntary cough strength on a scale of 0 to 5, and the magnitude of endotracheal secretions from none, mild, moderate, to large. The semiobjective scale of cough strength used by the single observer was as follows: 0 = no cough on command, 1 = audible movement of air through the endotracheal tube but no audible cough, 2 = weakly (barely) audible cough, 3 = clearly audible cough, 4 = stronger cough, and 5 = multiple sequential strong coughs. For the purposes of this study, a more objective measure of cough, termed the white card test (WCT), was created. A single observer (M.K.) placed a white file card 1 to 2 cm from the end of the endotracheal tube and asked patients to cough, up to three to four times, just prior to endotracheal extubation. If any wetness appeared on the card, it was classified as a positive WCT result. The amount of secretions was assessed by this same single observer who classified patients as producing either no, mild, moderate, or abundant endotracheal secretions based on personal observations and collective information from the nursing and respiratory therapists from their assessments within the 4 to 6 h preceding endotracheal extubation. Therapists and the bedside nurses also reported the approximate frequency of endotracheal suctioning during this period prior to extubation.
Patients were then extubated as per the instructions of their attending
physicians. Patients who remained extubated at 72 h were
classified as having successful extubation. Since this study did not
stipulate criteria for reintubation (decisions were made by
caregivers), physiologic parameters were recorded prior to reintubation
that were thought to have contributed to the decision to reintubate.
These reasons were categorized into three general groups: hypoxemia,
defined as P:F ratio < 120 or pulse oximetry desaturation on a
nonrebreather, high-flow oxygen face mask; hypercapnia, if arterial
blood gas analysis revealed acute hypercapnic respiratory failure or
sustained respiratory rate 
35 breaths/min; and failure to maintain
an adequate airway for either mental status changes or inadequate
expectoration. Patients were followed for this study until
discharge from the ICU or until placement of a tracheostomy.
Statistical Analysis
Mean values of selected demographic variables and
physiologic parameters of patients who were successfully extubated were
compared by nonpaired Student’s t tests to those who could
not be extubated. For grouped data, 
2 and/or
Fisher’s Exact Test were used in comparing differences in proportions
between the two groups and in deriving p values. Cough strength and the
magnitude of secretions were reclassified as binary variables; patients
with absent or barely audible (grade 0 to 2) cough were compared to
those with clearly audible, moderate-to-strong (grade 3 to 5) cough,
and those with no or mild secretions were compared to those with
moderate-to-abundant secretions. Risk ratios (RRs) were computed as the
preferred measure of the strength of association between the parameters
of interest and extubation outcome. A multiple logistic regression
model was subsequently used to adjust for confounders, assess effect
modifiers, and identify parameters that independently predicted
extubation outcome. The parameters included in the logistic model were
determined by the biological plausibility of each variable and/or
evidence of an association in the univariate analysis. Where
appropriate, threshold levels were defined using standard norms or
mean/median values where no such norms exist. Based on the findings of
the multiple logistic model, stratified analyses were performed to
assess the interplay of the independent predictors and to compute
measures of biological interaction on an additive scale using the
indexes of Hogan et al5
and Rothman.6
In the
absence of any interaction, the combined effects of the factors would
be purely additive, yielding a Rothman synergy index (RSI) of one, and
a Hogan index (HI) of zero. For synergism, the combined effect
would exceed the sum of the net independent effects of the individual
factors yielding RSI >1 and HI >0. For antagonism between the
factors, the combined effects of the factors would be less than the sum
of their net independent effects. Thus, the RSI would be < 1 and the
HI would be negative (< 0). Also, sensitivity, specificity, and
predictive values of the various parameters were computed. A
true-positive test result was defined as one that predicted extubation
success and the extubation was successful. A true-negative test result
was defined as one that predicted extubation failure and extubation was
unsuccessful. A false-positive test result was one that predicted
success when the extubation was unsuccessful. A false-negative test
result was one that predicted failure and the extubation was
successful. A receiver operator characteristic curve was constructed
for the continuous variable found to be an independent predictor of
outcome. A p value < 0.05 was used to signify statistical
significance.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
10 g/dL) were more likely to fail
extubation after passing the SBT. Patients with weak or no cough were
four times as likely to have unsuccessful extubations compared to those
with clearly audible (moderate) or stronger coughs on command (RR, 4.0;
95% confidence interval [CI], 1.8 to 8.9). The results of the WCT
were wholly dependent on cough strength and were not influenced by the
amount of endotracheal secretions. Patients with a negative WCT result
were three times as likely to have unsuccessful extubations compared to
those with a positive WCT result (RR, 3.0; 95% CI, 1.3 to 6.7).
|
Patients with lower hemoglobin levels ( 10 g/dL) were about five
times as likely to have unsuccessful extubations compared to those with
higher levels (RR, 5.3; 95% CI, 2.1 to 13.6). Nonetheless, 20 of 33
(61%) with such hemoglobin levels were successfully extubated.
Neither the degree of oxygenation as indicated by P:F ratio nor the RSBI was significant in predicting extubation failure in this cohort. Figure 1 shows the dispersion of the P:F ratios (receiving mechanical ventilation) for the study cohort. Seventeen of 19 patients with P:F ratios from 120 to 200 were successfully extubated. There was no significant difference in P:F ratios between successes and failures, and the risk of failure was no greater in patients with P:F ratios from 120 to 200 compared to P:F ratios > 200 (RR, 0.5; 95% CI, 0.1 to 2.0).
|
65 years or 75 years), APACHE II ( 20), duration of
intubation (> 4 days or > 7 days), and presence of congestive heart
failure, coronary artery disease, or COPD were not independently
associated with extubation outcomes. Cough strength was weakly affected
by the presence of encephalopathy. Poor cough strength and abundant
secretions were synergistic in predicting extubation failure (Table 5
). Patients with weak or no cough on command and with moderate or
abundant secretions were 32 times as likely to have unsuccessful
extubation compared to those with moderate or stronger cough and scant
or no secretions. This was nearly four times the net additive effects
of moderate or abundant secretions and poor or absent cough in
predicting extubation failure (RSI, 3.7; HI, 0.57). Similar results
were obtained using suctioning frequency as the measure of secretions
or the WCT result as a measure of cough strength. In fact, whereas 9 of
13 patients (69%) who required endotracheal suctioning at or more
frequently than every 2 h and who had a weak cough failed
extubation, none of the 38 patients with less frequent suctioning and
strong cough had unsuccessful extubation. There was no additional
synergistic effect with low ( 10 g/dL) hemoglobin level.
|
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
10 g/dL) appears to play an independent
role in predicting failure. Traditional "weaning" parameters, such
as the RSBI and P:F ratio, were not good predictors of extubation
outcomes in this cohort of medical patients who had successfully
completed an SBT. That factors of airway competence should play such a decisive role in predicting extubation outcome is not surprising. Sustenance of good gas exchange following endotracheal extubation demands that patients maintain a patent native airway and continue to breathe without the aid of the ventilator. Excessive endotracheal secretions, especially in the absence of a good cough, could lead to bronchial plugging, atelectasis, and/or aspiration pneumonitis, all of which can cause respiratory failure. Also, given that many of these patients have marginal respiratory muscle reserve, respiratory failure unrelated to airway competence could occur as a consequence of increased work of breathing. Previous studies found one or more of these to be the reasons for reintubation in similar patients. Epstein and Ciubotaru7 reviewed the records of 74 medically ill patients who required reintubation in spite of having successfully completed SBTs prior to extubation: respiratory failure (n = 21), congestive heart failure (n = 17), excessive endotracheal secretions (n = 12), upper-airway obstruction (n = 11), and encephalopathy (n = 7) were the principal causes. In such a retrospective study, it is difficult to determine the relative contribution of airway incompetence to the pathogenesis of respiratory failure. A more recent study4 of brain-injured patients examined the effects of a number of clinical factors, including airway parameters, on extubation outcomes; although "many patients were successfully extubated despite poor markers of airway function and clearance," the presence of a spontaneous cough and a lower frequency of endotracheal suctioning on the extubation readiness day were associated with successful extubation. Our study extends the importance of "airway factors" on extubation outcomes to critically ill patients with a variety of medical illnesses. However, our study did not finely grade the degree of encephalopathy, another potential cofactor that could affect readiness for extubation, and the ability to cough on command.
The mechanism by which a low concentration of hemoglobin
increased the risk of extubation failure is not clear. In previous
studies, some patients with weaning failure experienced insufficient
global oxygen delivery8
or myocardial
ischemia,9
which anemia may exacerbate. However, we did
not collect hemodynamic, oxygen metabolism, or ECG data that would be
useful in examining potential mechanisms to explain this observation.
Moreover, the pathogenesis of weaning failure may differ significantly
from that of extubation failure. Nonetheless, it is noteworthy that 20
of 33 patients (61%) with hemoglobin levels 10 g/dL were
successfully extubated, and would not have been considered for SBTs or
extubation in some major clinical trials.10
11
12
Accordingly, anemia by itself should not be used to preclude patients
from consideration for SBTs or endotracheal extubation. Data from a
previous, large prospective study13
demonstrated no
benefit in arbitrary transfusions to hemoglobin levels > 10 g/dL. Our
data do not justify systematic transfusion of patients receiving
mechanical ventilation to this level, but a future study should be
performed to determine whether patients who initially do not
successfully complete SBTs or extubation benefit from a higher
concentration of circulating hemoglobin.
To our knowledge, this is the first study to examine the effects of oxygenation on extubation outcomes. Although some clinicians and investigators use a P:F ratio of > 150 to commence weaning, many still use a threshold of P:F ratio of > 200. Since our ICU uses a P:F ratio of > 120 as one criteria to begin SBTs, we had the opportunity to scrutinize the effects of oxygenation on extubation outcomes. Note that in our study all patients (19 of whom who had P:F ratios from 120 to 200) had successfully completed the SBT as a requirement for enrollment. Since we did not follow up patients who did not successfully complete their SBTs, our results do not address the effect of the P:F ratio on weaning outcomes. Nonetheless, the successful extubation of 17 of 19 patients with P:F ratios from 120 to 200 suggests that the oxygenation threshold (P:F ratio > 200) used to commence SBTs in some major studies10 14 15 may be unnecessarily stringent. The small sample (n = 5) with a P:F ratio from 120 to 150 precludes drawing definitive conclusions in this particularly hypoxemic subset, although four of five patients were successfully extubated.
Numerous studies1 2 3 have suggested that the RSBI is among the best available predictors of combined liberation/extubation outcomes in patients who have not yet successfully completed an SBT. The fact that the RSBI does not help distinguish extubation outcomes suggests, as one would expect, that it gauges the ability to breathe without the ventilator. In other words, the RSBI primarily predicts SBT outcome, which may have greater weight in determining combined outcomes than whether the artificial airway is still required.
The measures of secretions and cough on command used in this study were somewhat subjective. To the best of our knowledge, there is no validated measure of cough strength or endotracheal secretions in critically ill patients. The use of a single observer was likely to ensure relative consistency in the application of the classification system. Nonetheless, some misclassifications could have occurred. Such misclassification is likely to be random (nondifferential), since patients were classified before extubation and the classification status had no bearing on decisions of extubation or reintubation. Moreover, the use of only two categories in the analyses guarantees that any effect of such random misclassification would bias the results toward the null.16 17 For example, if one assumes that poor cough strength is truly associated with extubation failure, then patients with weak cough wrongly misclassified in the moderate-to-strong cough category would lead to overestimation of extubation failure in that group. Similarly, patients with strong cough wrongly misclassified in the weak cough group would lead to underestimation of extubation failure in that group. The true RR of extubation failure for patients with weak cough compared to those with strong cough would be higher than that reported in the presence of misclassifications. Hence, it is unlikely that the impressive findings of the study can be explained by such misclassifications. Replication of the study findings using the frequency of suctioning and the WCT as second, arguably more objective, measures of the amount of secretions and cough strength supports the important roles of these variables in predicting extubation outcomes. We suspect that many clinicians already integrate these (or similar) airway "parameters" into weaning and extubation decisions, but this is the first study to formally evaluate the predictive utility of these factors in this patient population. This and the study4 of brain-injured patients suggest that cough and secretions affect extubation outcomes. Further studies, preferably using more objective/validated measures and multiple observers, are required to generalize these findings. Finally, the current study is also limited by the fact that neither the decision to extubate nor the decision to reintubate was based on explicit, enforced protocols; physicians caring for the patients ultimately made the decisions. This further hinders the generalizability of this study and supports the need for future confirmatory studies.
If the importance of cough and secretions is confirmed, how could these be used to inform extubation decisions? One could argue that higher mortality associated with extubation failure18 demands that we minimize the number of unsuccessful extubations. No single measure of cough strength or secretions had sufficient sensitivity and specificity to be used, by itself, in making extubation decisions (Table 6) . However, patients with poor cough and moderate-to-abundant secretions were highly (82%) likely to have unsuccessful extubations despite successfully completing an SBT (Table 5) . Those with moderate-to-abundant secretions and good cough appear to represent a group at intermediate risk (19% had unsuccessful extubations), and clinicians should exercise caution in extubating such patients. Those with mild secretions (or who require endotracheal suctioning less frequently than every 2 h) and good cough are very unlikely to have unsuccessful extubations and can be extubated with reasonable confidence after successfully completing an SBT.
In conclusion, this study suggests that for patients who have
successfully completed an SBT, the magnitude of the cough on command
and the quantity of endotracheal secretions are significant predictors
of extubation outcomes. Extubation failure is synergistically enhanced
in the presence of abundant endotracheal secretions and absent or weak
cough. These factors of airway competence, or "extubation
parameters," may be more important than traditional "weaning
parameters," including the RSBI and P:F ratio, in predicting
extubation outcomes of patients who have successfully completed SBTs.
Further studies utilizing a greater number of observers and/or more
objective measures of cough and secretions are required. Finally, in
such a cohort, many patients with P:F ratio of < 200 (120 to 200) or
hemoglobin level of 10 g/dL, values traditionally used to preclude
weaning, may be extubated successfully.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Supported by an unrestricted research grant from the Daniell Family Foundation.
Received for publication September 20, 2000. Accepted for publication February 23, 2001.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Ferrer, M. Valencia, J. M. Nicolas, O. Bernadich, J. R. Badia, and A. Torres Early Noninvasive Ventilation Averts Extubation Failure in Patients at Risk: A Randomized Trial Am. J. Respir. Crit. Care Med., January 15, 2006; 173(2): 164 - 170. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Levy Pathophysiology of Oxygen Delivery in Respiratory Failure Chest, November 1, 2005; 128(5_suppl_2): 547S - 553S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Ouellette The Impact of Anemia in Patients With Respiratory Failure Chest, November 1, 2005; 128(5_suppl_2): 576S - 582S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Levy, E. Abraham, M. Zilberberg, and N. R. MacIntyre A Descriptive Evaluation of Transfusion Practices in Patients Receiving Mechanical Ventilation Chest, March 1, 2005; 127(3): 928 - 935. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Frass, C. Dielacher, M. Linkesch, C. Endler, I. Muchitsch, E. Schuster, and A. Kaye Influence of Potassium Dichromate on Tracheal Secretions in Critically Ill Patients Chest, March 1, 2005; 127(3): 936 - 941. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Sadowski, R. E. Dechert, K. P. Bandy, J. Juno, V. Bhatt-Mehta, J. R. Custer, F. W. Moler, and S. L. Bratton Continuous Quality Improvement: Reducing Unplanned Extubations in a Pediatric Intensive Care Unit Pediatrics, September 1, 2004; 114(3): 628 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Manthous, Y. Amoateng-Adjepong, H. E. Fessler, and J. A. Krishnan Weaning by Protocol Am. J. Respir. Crit. Care Med., July 1, 2004; 170(1): 98 - 100. [Full Text]
|
|
|
|
|
|
M. Smina, A. Salam, M. Khamiees, P. Gada, Y. Amoateng-Adjepong, and C. A. Manthous Cough Peak Flows and Extubation Outcomes Chest, July 1, 2003; 124(1): 262 - 268. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Manthous The Anarchy of Weaning Techniques Chest, June 1, 2002; 121(6): 1738 - 1740. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
