|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Commission dEpidémiologie et de Recherche Clinique of the Société de Réanimation de Langue Française, Villejuif, France.
A list of members of Commission dEpidémiologie et de Recherche Clinique is located in the Appendix.
Correspondence to: François Blot, MD, Service de Réanimation Polyvalente, Institut Gustave Roussy, 39 Rue Camille Desmoulins, 94805 Villejuif Cedex, France; e-mail: blot{at}igr.fr
| Abstract |
|---|
|
|
|---|
Design: A retrospective study.
Setting: A questionnaire was sent to all ICUs included in the national registry of the Société de Réanimation de Langue Française.
Patients: All ICU patients receiving mechanical ventilation (MV) during the year preceding the survey (excluding noninvasive ventilation) were covered by the questionnaire.
Intervention: None.
Measurements and results: Of the 708 ICUs invited to participate, 152 (21.5%) answered the questionnaire. Overall, 35,322 patients (median, 212 patients per unit; interquartile range [IQR], 148 to 329 patients) had received MV, including one fourth of patients for > 7 days. A median of 7.2% of patients (IQR, 2.9 to 11.1%) had undergone a tracheostomy. The most frequently reported indications for tracheostomy were prolonged MV (95%, after a median of 20 days) and failure of extubation (48%). An indication for "early" tracheostomy (ie, < 3 weeks) was considered in 68% of the ICUs, after a median time of 7 days. Physicians thought that performing tracheostomy would facilitate weaning from MV and early oral nutrition, and would improve overall patient comfort. The main disadvantages listed were tracheal complications, the aggressiveness of the procedure, and the risk of infection. Surgical techniques largely remained preferred compared to percutaneous techniques.
Conclusions: There is much heterogeneity in the reported practices of tracheostomy in French ICUs. Three fourths of physicians considered that a randomized study comparing prolonged translaryngeal intubation and early tracheostomy in patients expected to receive prolonged MV would be necessary to clarify those issues.
Key Words: ICU mechanical ventilation tracheostomy
| Introduction |
|---|
|
|
|---|
The aim of this descriptive study was to investigate the current practice of tracheostomy in French ICUs through a nationwide survey. The frequency, timing, indications, and techniques of tracheostomy were recorded.
| Materials and Methods |
|---|
|
|
|---|
The following items were recorded: hospital category (teaching hospital or not); the number of beds in the hospital and in the ICU; ICU category (ie, medical, surgical, or medical-surgical ICU); the number of patients who had received MV (excluding noninvasive ventilation) in the preceding year, with the distinction made between medical and surgical patients; the number of patients receiving MV for > 7 days; the number of patients undergoing a tracheostomy during their ICU stay; the techniques used to perform the tracheostomy (ie, surgical or percutaneous); the operator and location (ie, surgeon or intensivist, in the operating room or the ICU); indications (no specific indication; failure of extubation followed by reintubation; prolonged MV, and if so, after how long; timing of tracheostomy after the initiation of MV and whether tracheostomy was performed early [ie, before receiving MV for 3 weeks] in specific indications).
In addition, senior physicians were asked to provide their opinions on the potential advantages and disadvantages of tracheostomy. For each item, a score between 1 (minor advantage or disadvantage) and 4 (major advantage or disadvantage) was given.
Finally, physicians were asked whether they considered that a prospective randomized trial comparing early tracheostomy vs prolonged intubation would be justified. The results are reported as the percentage or median and interquartile range (IQR).
| Results |
|---|
|
|
|---|
Type and Size of Participating ICUs
Of 708 ICUs that were invited to participate, 152 (21.5%) answered the questionnaire. The majority of participating ICUs were medical-surgical ICUs (77%), 15.8% were medical ICUs, and 7.2% were surgical ICUs. Approximately one third of ICUs (38%) were located in teaching hospitals. The median number of beds in participating ICUs was 12 (IQR, 8 to 15 beds).
The nonresponding ICUs, which were included in the registry of the national society, were compared to the 152 ICUs who answered the questionnaire. The number of beds per ICU was comparable between the ICU groups (median, 10 beds; IQR, 8 to 12 beds), as was the percentage of medical ICUs (17.2%). However, a lesser proportion of nonresponding ICUs was located in teaching hospitals (25%; p < .05 [
2 test]).
Patients Receiving MV During the Previous Year
The median number of patients receiving MV per unit (excluding noninvasive MV) during the year preceding the study was 212 (IQR, 148 to 329 patients; total, 35,322 patients). A majority of patients had received MV for a medical indication (median, 125 patients; IQR, 82 to 200 patients; total, 16,594 patients). Overall, 9,059 patients (median, 54 patients per unit; IQR, 35 to 108 patients per unit) received MV for > 7days. The number of patients receiving MV per year was slightly higher in teaching hospitals and in surgical ICUs than in other units. Otherwise, there was no significant difference between teaching hospitals and others, or among medical-surgical ICUs, surgical ICUs, and medical ICUs (Table 1
).
|
Physicians Opinion Survey on Indications and Timing of Tracheostomy
The most frequently reported indications for tracheostomy (multiple answers were allowed, and ear, nose, and throat [ENT] emergencies were excluded) were failure of extubation (as reported by 48% of responding physicians) and prolonged MV (as reported by 95% of physicians) after a median period of 20 days (IQR, 14 to 30 days). The absence of any specific indication was notified by 3% of responding physicians. The failure of extubation was more often considered as an indication for tracheostomy in nonteaching hospitals (54%, after a median time of 15 days) than in teaching hospitals (38%, after a median time of 21 days).
Early tracheostomy (before 3 weeks of MV) was considered by 68% of the respondents, after a median time of 7 days (IQR, 3 to 8 days). Early tracheostomy was considered more often in nonteaching hospitals (76.5%) than in teaching hospitals (54.5%; (p < 0.01 [
2 test]) after a median time of 7 days in both types of hospitals. Only 58% of medical ICU physicians considered performing an early tracheostomy, compared to 70 to 72% of physicians in other ICUs.
The potential advantages and disadvantages of tracheostomy listed by physicians are shown Table 2 . Briefly, the main advantages listed were as follows: (1) facilitation of weaning, (2) overall comfort of patients, and (3) earlier oral nutrition. The main disadvantages were as follows: (1) tracheal complications, (2) aggressiveness of the procedure, and (3) risk of stomal infections. A prospective randomized study comparing prolonged translaryngeal intubation and early tracheostomy in patients receiving long-term MV was considered to be justified by 73% of physicians.
|
|
| Discussion |
|---|
|
|
|---|
Despite the frequent use of tracheostomy in ICUs, there is little epidemiologic information on this intervention. In this multicenter survey, French ICU senior physicians were queried on their practices concerning the performance of tracheostomy in patients receiving MV. The first limitation of this work is that only one fifth of ICUs invited to participate actually answered the questionnaire. Although the participating ICUs were roughly comparable to the nonresponding ones, the generalization of our results is questionable. Specifically, physicians answering the questionnaire could be those having a greater interest in the practice of tracheostomy. A second limitation is that the study is based in part on self-reporting by physicians, in addition to objectively recorded data. Finally, the role of tracheostomy within a strategy of weaning from MV, and, more specifically, with regard to the use of noninvasive MV after extubation, has not been investigated in the present study, although its role might be growing, as some studies89 have suggested that postextubation noninvasive ventilation in patients who are at risk for respiratory failure does not prevent reintubation.
About one fourth of patients receiving MV underwent MV for >7 days, and about one third of them were reported to have received a tracheostomy (7.7% of all patients receiving MV). In a 1-day international point-prevalence study,6 24% of patients received MV via a tracheostomy. In Swiss ICUs, 1.3% of all patients and 10% of those receiving MV for > 24 h eventually received a tracheotomy.7 In three North-American studies,101112 8.3 to 10% of about 6,500 patients receiving MV required a tracheostomy. Although these other studies used different designs and methods, which does not allow direct comparisons among them, physicians practices appear to vary considerably from one country to another. In particular, the prevalence of tracheostomy seems to be higher in France compared with that in other European countries.
The opinions of French physicians about the indications for tracheostomy were similar to those reported in the Swiss study.7 The need for long-term MV was the major indication, followed by failed weaning and reintubation. Overall, tracheostomies were said to be performed after a median time of 20 days, a duration that is longer than that reported in the international survey by Esteban et al6 (median, 11 days), in the Swiss study7 (second week of MV), or in the recent study by Engoren et al12 (median, 13 to 17 days).
However, an early tracheostomy appears to be considered in two thirds of the ICUs, after a mean of 7 days. Recently, the VIII Consensus Conference of the French Society of Intensive Care Medicine, which focused on artificial airways, suggested13 that a decision on performing tracheostomy could be made without delay after 5 to 7 days of MV, if at that time the foreseeable duration of MV exceeds 15 additional days. In 1989, the North-American Consensus Conference recommended14 that translaryngeal intubation should be preferred in the first 10 days of MV, while tracheostomy should be performed when MV exceeds 3 weeks. Both recommendations therefore suggest performing a tracheostomy when the expected duration of MV is
3 weeks. However, the major difficulty in applying a recommendation for early tracheostomy is to predict the duration of MV. Criteria predictive of prolonged MV or of extubation failure have been described in several patient populations.1015161718 However, most of these criteria lack predictive value at the individual level. Although the ideal timing of tracheostomy remains undetermined,19 early tracheostomy could be more particularly recommended for some indications, such as cervical spinal cord injury or infratentorial lesions.17
Physicians responding to this survey identified several advantages and disadvantages of tracheostomy (Table 2). The potential advantages recorded (eg, facilitation of weaning, better patient comfort, earlier oral nutrition and speech, easier care of airways, and safety) have been widely discussed in the literature.202122232425262728 In an international multicenter study,29 373 patients with tracheostomy were compared to a matched control group. Although the mortality rate in the ICU was lower in tracheostomized patients (24.4% vs 37.3%, respectively; p < 0.001), the hospital mortality rate did not differ between the two groups (38.6% vs 41.6%, respectively; p = 0.41). Tracheostomy can substantially reduce the mechanical workload of ventilator-dependent patients.4 Except for the recent study by Rumbak et al,5 other studies2930 have not documented the other potential advantages of the procedure, such as a reduced rate of sinusitis or ventilator-associated pneumonia, or a reduced duration of MV or ICU and hospital stays, and better comfort for the patient.
Similarly, the main disadvantages that have classically been advocated (eg, tracheal complications, catheter-related infections, and psychological consequences) cannot be specifically attributed to tracheostomy. Interestingly, the hospital policy to accept or refuse tracheostomized patients on regular wards or the availability of beds for long-term tracheostomized and ventilator-dependent patients has opposite consequences on ICU practices. Fifteen physicians considered that performing a tracheostomy may facilitate ICU discharge, while 11 others considered that tracheostomy was a potential barrier to ICU discharge. Finally, absolute contraindications for tracheostomy are altogether infrequent. Anatomic defects or soft tissue infection of the neck, or very severe respiratory distress with refractory hypoxemia and hypercapnia may be considered as contraindications. Conversely, although hematologic disorders are often considered to be contraindications for the procedure, previous studies3132 have shown that a tracheostomy can be safely performed in patients with severe neutropenia or thrombocytopenia.
Whereas the indications and potential benefits of tracheostomy remain highly controversial, an extensive number of studies on the techniques and procedures of tracheostomy has been published during the past 20 years. Despite the low rates of complications associated with tracheostomy that have been reported over the past few years in the literature,3334 most physicians responding to this survey still consider the procedure to be aggressive. Most probably, they simply bear in mind that it is an invasive surgical procedure, which may expose patients to the risk of cosmetic sequelae. The development of percutaneous techniques could change the approach of intensivists regarding the use and timing of tracheostomy. In our survey, tracheostomies were indifferently performed by a surgeon/ENT specialist or by the intensivist, in the operating room or in the ICU. Surprisingly, despite a promising increase in the number of studies in the literature on percutaneous techniques,3334 surgical techniques remain largely preferred in France (always used, 84%; often used, 20.5%). Unfortunately, ICU physicians were not queried about the reasons for their reluctance to use percutaneous techniques. Similarly to Swiss ICUs, organizational aspects and lack of adequate training in the use of these techniques could, in part, explain this finding.
We conclude that in the 152 French ICUs that answered the questionnaire, there is a great heterogeneity regarding the reported routine practice of tracheostomy in patients receiving MV. The need for long-term MV and the failure of extubation/reintubation remain the major indications for performing tracheostomy in the opinions of French intensivists. In this setting, tracheostomy is most often considered after a mean time of 3 weeks (ie, later than recommended by several consensus conferences).1314 As suggested by three fourths of the physicians answering the questionnaire, a randomized study comparing prolonged translaryngeal intubation and early tracheostomy in patients who are expected to receive prolonged MV would be necessary to clarify those issues.
| Appendix |
|---|
|
|
|---|
| Footnotes |
|---|
Received for publication July 8, 2004. Accepted for publication October 26, 2004.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |