Do Arterial Catheters Improve Patient Care in the ICU?

By David J. Pierson, MD, Editor

SYNOPSIS: This large propensity-matched cohort analysis of critically ill, mechanically ventilated patients in 139 U.S. ICUs found no evidence that the use of arterial catheters improved patient outcomes.

SOURCE: Gershengorn HB, et al. Association between arterial catheter use and hospital mortality in intensive care units. JAMA Intern Med 2014 Sep 8. [Epub ahead of print].

Gershengorn and colleagues used prospectively collected data from the Project IMPACT database (a nationwide, voluntary, proprietary database for assessing performance of U.S. ICUs with respect to patient outcomes and numerous other variables) to examine the question of whether the use of arterial catheters (ACs) in critically ill, mechanically ventilated patients was associated with improved survival or other documentable benefits. The data used were gathered from 2001 through 2008 in 139 ICUs in 119 hospitals, and included 60,975 adult patients. The investigators used propensity score matching of patients with and without ACs in an attempt to eliminate differences other than the use of the catheters.

A propensity score is a measure reflecting the propensity of a patient, based on other characteristics, to receive a particular intervention, in this case placement of an AC. Propensity matching is a technique used with a retrospective patient cohort to match individuals who received the intervention (an AC) with those who did not but had the same propensity to do so, creating a case-control study. With hospital mortality as the main outcome variable, the authors examined the overall cohort of mechanically ventilated patients by four alternative methods of comparison, and also investigated nine secondary cohorts in the same ICUs to assess the generalizability of the findings.

Propensity score matching yielded 13,603 pairs of patients who did and did not receive an AC. The patients were critically ill, with 63.5% mechanically ventilated on ICU admission and 44.5% requiring vasopressors. Hospital mortality in both groups was approximately 36%. Among the patients, 73% of the patients had central venous catheters and 5% had pulmonary artery catheters (PACs). By multiple statistical techniques, there was no association between AC use and hospital mortality in the overall cohort (odds ratio for not having an AC, 0.98; 95% CI 0.93-1.03). In eight of the nine secondary cohorts, there was no association between AC use and hospital mortality by propensity-matched analysis; in the secondary cohort of patients requiring vasopressors, the odds of death were increased in patients who received an AC (OR, 1.08; 95% CI, 1.02-1.14; P = 0.008). From this retrospective study, the authors concluded that ACs do not improve the ability to care for critically ill ICU patients and could cause harm. They call for prospective studies to validate the associations found and assess causation.


This study is reminiscent of the landmark study by Connors et al 20 years ago1 that first documented the lack of outcome benefit from the use of PACs in critically ill patients. As Gershengorn and colleagues note, the Connors study — an association study like this one that could not establish causality — ushered in a series of randomized controlled trials that ultimately confirmed the findings that PACs do not, in and of themselves, improve patients’ likelihood of survival from critical illness. The magnitude of practice change resulting from those trials is illustrated by the finding that only one patient in 20 in the present study got a PAC; 20 years ago, most if not all of them would probably have had a PAC in addition to an AC.

If the Gershengorn study is followed by prospective trials of the outcome benefits from ACs (a substantial "if," given the current funding environment and the more modest commercial implications of ACs as compared to PACs), and if they confirm the present findings, it may be that ACs will be as uncommon in our ICUs as PACs a couple of decades from now.


  1. Connors AF Jr, et al; SUPPORT Investigators. The effectiveness of right heart catheterization in the initial care of critically ill patients. JAMA 1996;276:889-897.