|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Pulmonary artery catheterization has evoked more controversy than any other widely adopted cardiovascular monitoring practice because it is an expensive, invasive technique that is widely used but still not rigorously
In 2003, the American Society of Anesthesiologists published an updated practice guideline for pulmonary artery catheterization that provided a current exhaustive review of the scientific evidence for the clinical effectiveness of PAC monitoring.[318] The inconsistent findings of the many studies in the literature were striking, with evidence both supporting[540] [541] [542] [543] [544] [545] and refuting [323] [324] [546] [547] [548] [549] [550] [551] [552] the benefits of this technique. Many early studies reported small numbers of patients, which limited their validity and wider applicability, and larger more recent trials continue to have mixed results and limitations in study design. One of the earliest observational studies involving more than 1000 general surgical patients reported lower perioperative mortality and myocardial reinfarction rates in patients monitored with PACs versus historical controls,[553] but other large studies have not found a similar outcome benefit. One reason for these divergent results may be that clinical trials have focused on different populations, such as patients with acute myocardial infarction,[554] [555] those undergoing cardiac surgery,[556] [557] and critically ill patients with a variety of medical and surgical diseases.[322] [558] Despite the large numbers of patients in some of these reports, many experts have questioned these studies because they were observational in design rather than randomized controlled trials.[318] [349] [559] [560] Nonetheless, several of these studies have raised serious concern among physicians because they suggested that not only was the PAC ineffective, but patients who received PAC monitoring also suffered increased morbidity and mortality when compared with patients who did not receive such monitoring.[322] [554] [555]
Perhaps the most provocative of all these PAC outcome studies is one published in 1996, in which Connors and colleagues examined the association between PAC use during the first 24 hours of intensive care and subsequent survival. [322] This prospective cohort study included 5735 patients in five teaching hospitals as part of a larger 9000-patient study entitled "The study to understand prognoses and preferences for outcomes and risks of treatments (SUPPORT)."[561] [562] [563] [564] The patients in this study were desperately ill; entry into the study required a predicted 6-month mortality rate in excess of 50%. Patients who underwent PAC monitoring were judged to be sicker by every measurement recorded, but the authors used case-matching analyses and applied a propensity score to adjust for these confounding medical covariates. To the extent that these statistical adjustments were successful, the results of the study were very worrisome. PAC-monitored patients had increased mortality, increased length of hospital stay, and increased cost. Moreover, no subgroup of patients appeared to benefit from PAC monitoring. Publication of this study was accompanied by a strongly worded editorial calling for a moratorium on PAC use or a randomized controlled trial to define its efficacy.[529]
The controversies surrounding PAC use have helped identify several key issues that remain unanswered but are intimately related to the basic question of PAC monitoring and patient outcome.
Tremendous variability in the level of skill, knowledge, and confidence is found in nurses and physicians who use PACs.[412] [413] [414] [415] [417] [418] Furthermore, experienced clinicians often choose different therapies when presented with the same PAC-derived hemodynamic data.[409] [410] [411] Some experts believe that failure to control for these factors has been responsible for the poor performance of the PAC in many of the clinical trials reported in the literature.[325] [559] [563] [565] As noted by Fowler and Cook, "A [research study] design choice not to standardize treatment but to replicate day-to-day care in ICUs around the world may reflect the reality that there is nothing standard about 'standard practice,' and there is no proven 'best practice.'"[566] Consequently, because many studies have not provided an explicit description of the level of knowledge of the practitioners or how the PAC was used to guide care, the results of these studies have been called into question.[318] [325]
The use of PACs varies widely between individual physicians, institutions, and geographic locations.[375] [558] [567] [568] As in other parts of modern medicine, this begs the question regarding whether the PAC is overused in some settings or perhaps underused in others. Some authors have suggested that PAC use has been driven in part by the variation in regional medical care payment structures or the local mechanism for physician reimbursement.[349] [567]
Learning bias may be another important factor that confounds determination of PAC effectiveness.[325] [349] This phenomenon is well recognized in clinical medicine. In brief, it suggests that we now take better care of patients in general because we have learned such a great deal about cardiovascular pathophysiology and patterns of disease by using PACs that we now apply this knowledge effectively in the care of patients who do not have PAC monitoring.
Early use of PAC monitoring in the care of general surgical patients, [569] [570] trauma patients,[544] cardiac surgical patients,[571] or elderly high-risk patients[541] has been emphasized by some authors as a requirement for improvement in clinical outcome. The studies of Shoemaker and coworkers in particular have underscored the importance of early monitoring and treatment to optimize tissue perfusion
The role of PAC monitoring in improving patient outcome is inextricably tied to the therapies that it guides. As several authors have noted, "if there is no good treatment for the pathophysiologic state that is identified, there is no expectation of therapeutic benefit from the diagnostic procedure."[334] [349] This may be one of the key issues in our current conundrum, namely, our inability to effectively treat many serious illnesses, such as sepsis, even when the cardiovascular problems are clearly identified through PAC monitoring.
In terms of PAC-guided therapies, one of the most controversial is one in which the physician measures hemodynamic variables with the catheter and then attempts to increase oxygen delivery to some predetermined goal. This treatment algorithm has been termed "maximizing oxygen delivery," "goal-oriented therapy," or "supraphysiologic goal-driven treatment." In essence, these therapies involve PAC monitoring to guide fluid and inotropic drug administration, with various supranormal circulatory goals as end points: oxygen delivery greater than 600 mL/min/m2 , oxygen consumption greater than 170 mL/min/m2 , or cardiac index higher than 4.5 L/min/m2 . Although some investigations have shown an outcome benefit with this therapeutic approach,[545] [569] [571] [574] [575] [576] [577] [578] others have shown either no effect of such therapy[579] or even increased mortality.[580] The uncertain efficacy of this therapeutic approach has been cited as a significant reason why PAC monitoring has not proved effective.[322] [559] A complete consideration of this treatment strategy is beyond the scope of this discussion, but it remains an extremely controversial aspect of PAC monitoring.[581] [582] Although most experts agree that PACs allow more accurate titration of traditional therapy in critically ill patients, some of the most "aggressive" therapies that it guides remain unproven.[583]
|
|
|
|
|
|
|
|
|
|
|
|
|
