Controversy: whether or not the use of a pulmonary artery catheter provides enough clinically-relevant information to justify the risks associated with its use, and if so, whether or not certain patient populations are more or less likely to derive benefit from this technology.
Data in Favor:
Data out of Favor:
According to a recent Cochrane Database Review, there are 12 available randomized studies on PA catheters, some of which are reviewed here. Keep in mind that, as mortality rates are low, most of the studies of PA catheters are underpowered to detect a mortality difference – in fact, the Cochrane database review found that only one of 12 studies was sufficiently powered.
Bender JS et. al. Ann Surg. 226: 229, 1997 (104 patients)
104 consecutive patients randomized to have a PAC placed the morning of surgery (group I) or only if clinically indicated (group II). Group I patients were resuscitated to preestablished endpoints before surgery and kept at these points both intraoperatively and postoperatively. Group II patients received standard care. Results: one death in each group, 13 intraoperative or postoperative complications in group I versus 7 in group II (NS). Group I patients received more fluid than did group II patients (5.1L vs. 3.8L, p < 0.003), no difference in intensive care unit LOS.
Rhodes A et. al. Intensive Care Med 28: 256, 2002 (201 patients)
Prospective, controlled, trial of 201 intensive care unit patients randomised either to a PAC group (n=95) or the control group (n=106). Results: no significant difference in mortality between the PAC and control groups (47.9% vs. 47.6%, p>0.99). The PAC group had significantly more fluids in the first 24h and an increased incidence of renal failure (35% versus 20%, p<0.05) and thrombocytopenia (p<0.03).
Bonazzi et. al. Eur J Vasc Endovasc Surg 23: 445, 2002 (100 patients)
A consecutive series of 100 patients randomised to either haemodynamic optimisation through the use of a pulmonary artery catheter (CI > 3.0 l/min/sqm, PWP > 10 and <18 mmHg, SVR <1450 dyne/sec/cm(-5), DO(2)> 600 ml/min/sqm) or conventional treatment. Results: no differences in in-hospital mortality, cardiovascular morbidity, postoperative renal failure or duration of hospital stay between the groups.
Sandham et. al. N Engl J Med 348: 5, 2003 (1994 patients)
Randomized trial of high-risk patients 60 years of age or older, ASA class III or IV, who were scheduled for urgent or elective major surgery, followed by a stay in an intensive care unit, PA catheter vs. none. Of 3803 eligible patients, 1994underwent randomization. Mortality was 7.8% in the PAC group, vs. 7.7% in the non-PAC group. 8 PEs in the PAC group, vs. none in the non-PAC group (p=0.004). Median hospital stay was 10 days in each group.
Warszawski J et. al. JAMA 290: 2713, 2003 (676 patients)
Randomized controlled study of 676 patients who fulfilled the criteria for shock, ARDS, or both conducted in 36 French ICUs. Patients randomly assigned to PCA (n = 335) or not (n = 341). Treatment was left to the discretion of each individual physician. Results: no significant differences in 14 day mortality (49.9% with PAC, 51.3% without, p =0.70), or mortality at 28 (p = 0.67) or 90 days (p = 0.71).
Harvey S et. al. Lancet 366: 472, 2005 (1041 patients)
Randomised controlled trial of 1041 patients from 65 UK ICUs, assigned to management with (n=519) or without (n=522) a PAC. Results: no difference in hospital mortality (68% vs 66%, p=0.39). There were 46 complications associated with insertion of a PAC, none of which was fatal.
Binanay C et. al. JAMA 294): 1625, 2005 (433 patients)
Randomized controlled trial of 433 patients at 26 sites randomized to receive therapy guided by clinical assessment and a PAC or clinical assessment alone. The target in both groups was resolution of clinical congestion, with additional PAC targets of a pulmonary capillary wedge pressure of 15 mm Hg and a right atrial pressure of 8 mm Hg. Medications were not specified, but inotrope use was explicitly discouraged. Results: PAC did not significantly affect the primary end point of days alive and out of the hospital during the first 6 months (p = 0.99), mortality (p = 0.35), or number of days hospitalized (p = 0.67). In-hospital adverse events were more common among patients in the PAC group (21.9% vs. 11.5%, p = 0.04), although there were no deaths related to PAC use.
Harvey S et. al. Cochrane Database of Systematic Reviews, Issue 4, 2008
Identified 12 total studies, two of which were multi-center and only one of which was adequately powered. “Efficacy studies are needed to determine optimal management protocols and patient groups who could benefit from management with a PAC.” The pooled OR for mortality in general ICU patients (8 studies) was 1.05 (CI 0.87-1.26), and the pooled OR for mortality in high-risk surgery patients (4 studies) was 0.99 (CI 0.73-1.24).