PREDICTING ICU LENGTH OF STAY FOLLOWING SINGLE-LUNG TRANSPLANTATION

Citation
Kh. Lee et al., PREDICTING ICU LENGTH OF STAY FOLLOWING SINGLE-LUNG TRANSPLANTATION, Chest, 110(4), 1996, pp. 1014-1017
Citations number
14
Categorie Soggetti
Respiratory System
Journal title
ChestACNP
ISSN journal
00123692
Volume
110
Issue
4
Year of publication
1996
Pages
1014 - 1017
Database
ISI
SICI code
0012-3692(1996)110:4<1014:PILOSF>2.0.ZU;2-H
Abstract
Study objective: The aim was to identify potential predictors of ICU l ength of stay (LOS) for single lung transplant patients. Design: Retro spective chart review. Setting: University medical center. Patients: A ll single lung transplant recipients for 1992 and 1993 at our institut ion. Results: Data were collected from 69 patients. The median ICU LOS was 5 days, and this was highly correlated with the duration of mecha nical ventilation. The mean acute physiology and chronic health evalua tion (APACHE II) score was 10. Patients with pulmonary hypertension ha d the longest ICU LOS. Similarly, patients with a measured transpulmon ary gradient of 20 mm Hg or less had a significantly shorter ICU LOS. Patients with an immediate postoperative PaO2/fraction of inspired oxy gen (FIo(2)) ratio greater than 200 mm Hg and a flow mismatch between the two lungs of 30% or less also had a significantly shorter ICU LOS. Positive and negative predictive values for the immediate postoperati ve PaO2/FIo(2) ratio of 200 mm Hg or less mere 77% for an ICU LOS grea ter than 5 days, and the calculated receiver operating characteristic (ROC) curve area was 0.74. Conclusion: Overall, the immediate postoper ative PaO2/FIo(2) ratio of 200 mm Hg or less had the best positive and negative predictive values as well as the highest ROC curve area for predicting an ICU LOS greater than 5 days.