| 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 |
|||||||||
Erie, PA
Dr. Chinsky practices with Chest Diseases of Northwestern Pennsylvania.
Correspondence to: Kenneth D. Chinsky, MD, FCCP, 3580 Peach St, Suite 103A, Erie, PA 16508; e-mail: LChinsky{at}aol.com
Tracheostomy has been described as far back as ancient Egypt.1 Pundits might suggest that physicians have been debating the indications, timing, and techniques ever since. The classic surgical approach for tracheostomy was demonstrated by Jackson2 almost 100 years ago. Although percutaneous tracheostomy has gained popularity over the last 10 years or so, it was actually first reported in 1957.3 Despite the long history of the procedure, the jury remains out as to the optimal approach.
The study by Blot et al in the current issue of CHEST (see page 1347) is the most recent attempt to address tracheostomy practices, this time comparing French ICUs. It is a retrospective descriptive study demonstrating the heterogeneity that is present in all aspects of the procedure. Although the data involve only a single country, I suspect that most practitioners will observe similar themes running through their own ICUs. As the authors point out, the 21.5% survey response rate questions whether the results are generalizable. Also, given that 25% of patients received ventilation for > 7 days, the population may not be representative of other ICUs. The failure of extubation is not well-defined, leaving open another source of variation. Furthermore, the patient population is not as well-described as I would have hoped. The reader does not know how many patients had neurosurgical diagnoses such as traumatic brain injury or what percent were considered terminally ill. These factors vary from unit to unit and may impact the tracheostomy rate. For example, patients with traumatic brain injury with a low Glasgow coma scale score may benefit from early tracheostomy.4 Finally, this study is unable to ask why these differences in tracheostomy rates and techniques exist.
Therein lies the rub. Studies involving tracheostomy often employ different design methods and patient groups, and are challenging to compare. As in the current study, tracheostomy practices are often self-reported rather than obtained from prospectively recorded databases, giving rise to a potential source of error. Although the decision to perform a tracheostomy is complex, it can be broken down into two simple questions concerning a calculation of the likely benefit for any individual patient and an estimate of the duration of mechanical ventilation.5 The different complications of translaryngeal intubation vs tracheostomy have been reviewed elsewhere in detail. Likewise, the potential advantages of tracheostomy, such as increased comfort, enhanced oral nutrition, and ease of ventilator weaning, have been extensively studied and will not be reiterated here.5 Nevertheless, objective proof of these potential advantages remains difficult to obtain.
Some studies have focused on patient-specific factors to identify those who might benefit from early tracheostomy. Kollef et al6 found that nosocomial pneumonia, witnessed aspiration, the need for aerosol treatments, and the need for reintubation were variables that were independently associated with tracheostomy. Other groups have studied certain diagnoses, and the need and timing of tracheostomy. Qureshi et al7 concluded that an aggressive approach was indicated in patients with infratentorial brain lesions requiring ventilatory support, given the high frequency of extubation failure. On the other hand, the routine performance of early tracheostomy in burn patients did not improve outcomes or result in earlier extubation.8 In a further attempt to identify objective criteria, Boynton et al9 thought that surgical patients with resolving respiratory failure who meet the criteria for a spontaneous breathing trial but are not successfully extubated in 24 h may benefit from early tracheostomy.
There are non–patient-related factors as well that impact on tracheostomy timing. Studies rarely clarify the age of the physicians performing the procedure or answering the questionnaires. In the 1960s, endotracheal tubes were rigid, led to mucosal damage, and tracheostomy was often recommended after 3 days in a ventilator-dependent patient. Laryngeal stenosis was a major concern as a complication of translaryngeal intubation. An influential study10 published in 1984 prompted many physicians to perform tracheostomy around or before the second week of intubation. Physicians trained during these eras may be guided by these data even though subsequent studies have not always confirmed similar results.11 Also, little is written comparing differences in tracheostomy practices among various specialties. My gut feeling (without objective data) is that surgeons, anesthesiologists, and critical care internists approach the decision-making process differently. Another non–patient-related factor that influences tracheostomy practices is related to where care for the tracheostomized patient is rendered. The availability of long-term acute care hospitals and nursing homes with ventilator units may not affect the overall duration of mechanical ventilation, but can decrease the length of stay in the ICU and indirectly influence studies about the benefits of tracheostomy. The availability of these facilities may prompt physicians to perform tracheostomy sooner in order to transfer patients from crowded ICUs. Finally, as an American physician who practices in a highly litigious state, I suspect that liability concerns would be more of an issue than the single physician noted in the study by Blot et al.
Similar controversies exist in regard to preferences of traditional surgical or percutaneous tracheostomy. Studies have attempted to compare the risks and benefits of the techniques, but they have varied in design and in the outcome measures assessed. For example, one controversial issue is whether or not percutaneous tracheostomy predisposes the patient to pneumonia.1213
Heffner5 said it best when he observed that tracheostomy is "a complex medical decision that defies simple solutions." He went on to say that patients my benefit from tracheostomy to different degrees at different times. The situation is analogous to the decision about the placement of a central venous catheter. There are general indications (eg, the use of pressors and the need for total parenteral nutrition) that most agree on and other "softer" indications such as ease of phlebotomy. Furthermore, there are different anatomic approaches (eg, subclavian, internal jugular, and femoral). There are advantages and disadvantages to each approach, such as less infectious risk with the subclavian approach, less pneumothorax risk with the internal jugular approach compared to that with the subclavian approach, and potentially fewer bleeding complications utilizing the femoral approach in patients with coagulopathy. For any individual patient, there are times when we choose one approach over another.
Agreement about most medical decisions is easiest when we start at the extremes. Those with progressive or irreversible causes of respiratory failure will not benefit from weaning attempts and should have tracheostomy done early.5 Conversely, most physicians would not rush to perform a tracheostomy in a terminally ill patient who is expected to die in the immediate future, even if they meet other usual criteria.
The large gray zone in the middle is where most of the debate and variation in care occurs. We should use the available medical literature, focusing on specific diagnoses and patient factors, to make individual decisions about the timing of tracheostomy. Clearly, we should continue our search for evidence-based medicine to document objective proof of the purported advantages and disadvantages, as outlined in Table 2 of the study by Blot et al. Because there are an infinite number of combinations of these factors, no single study can dictate when the procedure should be done for any single patient. Similar to the central venous catheter analogy, there are instances in which the traditional approach is preferable to the percutaneous approach and times at which we wait 21 days to perform tracheostomy as opposed to an earlier time.
Indeed, the trend does seem to be toward the earlier performance of tracheostomy. A study14 of North Carolina hospitals from 1993 to 2002 showed that physicians ordered tracheostomy 2 days earlier than they did 10 years ago. Simple-to-follow decision trees guiding this process have been published.15
The major challenge in writing a medical editorial is not simply in critiquing a specific study, but in integrating the data into existing medical literature, determining what new information it contributes, and attempting to hypothesize new ideas or theories. With this in mind, I wonder whether tracheostomy practices are truly changing. Over 80 years ago, Jackson16 stated that "we all preach doing it [tracheostomy] early; but almost always do it late." Only time, and surveys such as that by Blot et al, will prove this point.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
