Table 1

Summary of clinical studies of PAC use
Ref.
 Year
 Number of cases
 Study design
 Clinical settings
 Significant findings
[41]
 1975
 413
 Case series
 Autopsy reports
 TEV with PAC was 4.25 times more frequent than with central lines; impact on mortality not studied
[38]
 1979
 116
 Prospective case series
 Critically ill patients with shock, pulmonary edema, and hemodynamic instability postoperatively
 77% Arrhythmia without increased mortality or morbidity, 1.7% staphylococcal bacteremia, 1.7% subclavian DVT
[49]
 1981
 60
 Incidence study
 Critically ill patients
 48% PVC and 33% VT with one death
[40]
 1981
 320 PAC in 219 patients
 Prospective case series
 Critically ill patients
 3% Major complications; only one death
[36]
 1983
 528 PAC placements in 500 patients
 Case series
 All cases in one medical center
 24% Complications, with 4.4% serious ones; no deaths related to complications
[42]
 1983
 36
 Prospective case series
 Autopsy
 61% mural thrombosis; incidence increased with prolonged duration of catheter; no significant impact on clinical course
[46]
 1984
 55
 Case series
 Autopsy, patients with PAC within 1 month of death
 53% RH endocardial lesions, 7% infective endocarditis, with pulmonic valve (56%) and pulmonary artery (5%) being the most and least common sites, respectively
[48]
 1985
 56
 Prospective case series
 ICU patents with shock, ARDS and preoperative
 12.5% advanced ventricular arrhythmia; no treatment required
[37]
 1985
 141
 Case series
 Autopsy
 PAC associated with higher rate of mural thrombi compared with central lines
[3]
 1987
 3263
 Retrospective
 Patients with acute myocardial infarction
 Increased length of hospital stay associated with PAC use; no long-term benefit
[33]
 1988
 88 (30/28/30)
 RCT (PAC control versus supranormal DO2 versus CVP)
 Preoperative high-risk surgical patients
 PAC had no effect on outcome unless used to guide therapy
[47]
 1989
 279 PAC
 Prospective
 ICU patients
 3% new RBBB
[14]
 1989
 1094 (537/557)
 Controlled prospective cohort
 Elective coronary artery bypass graft
 No significant difference in outcome between PAC and CVP groups
[4]
 1990
 5841
 Retrospective, analysis of PAC registry
 Patients with acute myocardial infarction
 Higher in-hospital mortality in CHF patients; thought to be related to use of PAC in sicker patients
[17]
 1991
 33 (16/17)
 RCT (PAC versus no PAC)
Nonsignificant benefit in favor of not receiving PAC
[43]
 1991
 297
 Prospective, incidence study
 Medical/surgical ICU
 22% local infection and 0.7% bacteremia; factors associated with high-risk catheter-related infection included skin colonization, IJ insertion, catheter placement >3 days and insertion in the OR
[28]
 1994
 100 (50/50)
 RCT (supra-normal DO2 versus normal DO2)
 Severe circulatory shock without response to fluid challenge
 Increase mortality in treatment group
[29]
 1995
 762 (252/253/257)
 RCT (control versus supranormal DO2 versus minimal SvO2)
 Multicenter, high-risk surgical patients with hemorrhagic, septic ARDS and trauma
 No difference in mortality, organ dysfunction, or length of stay
[45]
 1995
 32442
 Retrospective chart review
 OR and ICU
 0.03% PA rupture with 70% mortality rate
[39]
 1995
 630 PAC placements in 118 patients
 Retrospective analysis
 Patients with aneurysmal subarachnoid hemorrhage
 13% catheter related sepsis, 2% CHF, 1.2% DVT, 1% pneumothorax; no PA rupture
[5]
 1996
 2016 (1008/1008)
 Prospective cohort, case matching analysis
 Critically ill patients
 Increased mortality, cost of care and length of ICU stay in PAC group
[22]
 1997
 104 (51/53)
 RCT (routine PAC versus clinically indicated PAC)
 Low-risk elective abdominal vascular surgery
 Routine PAC had no benefit in mortality or morbidity
[21]
 1998
 120 (60/60)
 RCT (PAC versus no PAC)
 Surgical low-risk AAA repair
 No benefit, possibly with higher intraoperative complications
[6]
 2000
 10,217
 Retrospective database study
 Nonoperative patients in medical and surgical ICU
 Direct association of PAC use with admission in surgical ICU, white race, care given by nonintensivist, and having private insurance
[8]
 2001
 4059 (221/3838)
 Prospective, observational cohort
 Elective major noncardiac surgery
 Increase in cardiac and noncardiac events with PAC
[18]
 2003
 1994 (997/997)
 RCT (PAC versus no PAC)
 High risk, >6-year-old surgical patients
 No benefit in PAC group, higher PE in catheter group, survival rate favored non-PAC group
[20]
 2003
 676 (335/341)
 RCT (PAC versus no PAC)
 Multicenter; shock and ARDS patients
 No impact of PAC on mortality or morbidity
[23]
 2005
 1041 (519/522)
 RCT (PAC versus no PAC)
 Multi-center, all adult ICUs
 No evidence of benefit or hospital mortality, 10% complications but not fatal
[24]
 2005
 433 (215/218)
 RCT (PAC versus no PAC)
 Multicenter, severely symptomatic CHF patients
 No evidence of benefit or overall mortality, 5% complications but none fatal

AAA, abdominal aortic aneurysm; ARDS, acute respiratory distress syndrome; CHF, congestive heart failure; CVP, central venous pressure; DO2, oxygen delivery; DVT, deep venous thrombosis; ICU, intensive care unit; IJ, internal jugular; MI, myocardial infarction; OR, operating room; PA, pulmonary artery; PAC, pulmonary artery catheter; PE, pulmonary embolism; RHC, right heart catheterization; RH, right heart; RBBB, right bundle branch block; RCT, randomized clinical trial; SvO2, mixed venous oxygen saturation; TEV, thrombotic endocardial vegetation; VT, ventricular tachycardia.

Hadian and Pinsky Critical Care 2006 10(Suppl 3):S8   doi:10.1186/cc4834