Advanced Search

Home > Journals > The Journal of Cardiovascular Surgery > Past Issues > The Journal of Cardiovascular Surgery 2003 October;44(5) > The Journal of Cardiovascular Surgery 2003 October;44(5):591-6



A Journal on Cardiac, Vascular and Thoracic Surgery

Indexed/Abstracted in: BIOSIS Previews, Current Contents/Clinical Medicine, EMBASE, PubMed/MEDLINE, Science Citation Index Expanded (SciSearch), Scopus
Impact Factor 1,632

Frequency: Bi-Monthly

ISSN 0021-9509

Online ISSN 1827-191X


The Journal of Cardiovascular Surgery 2003 October;44(5):591-6



“Learning curves” of cardiac surgery in relation to risk stratification and hospital location

Anagnostopoulos C. E. 1,2,3, Siminelakis S. 1, Ananiadou O. 1, Katsaraki A. 1, Drossos G. 1 , Katritsis D. 2, Panagiotopoulos J. 2, Papadopoulos G. 1

1 Depart­ment of Car­diac Sur­gery, Uni­ver­sity of Ioan­nina, ­Greece
2 Depart­ment of Car­diac Sur­gery, Athens Euro­clinic, ­Greece
3 Sec­tion of Car­diac Sur­gery, St. ­Luke’s Roose­velt Hos­pital, ­Columbia Uni­ver­sity, New ­York, NY, USA

Aim. The pur­pose of ­this ­study was to deter­mine any sig­nif­i­cant dif­fer­ences in “­learning ­curves” ­between pri­vate and ­public hos­pi­tals ­when the ­same ­senior sur­geon was respon­sible ­during the ­initial ­phases of ­open-­heart sur­gery pro­grams devel­op­ment, in rela­tion to ­risk strat­ifi­ca­tion and hos­pital loca­tion.
­Methods. A pros­pec­tive ­review of 610 ­patients ­records was per­formed at a ­newly-­opened car­di­oth­o­racic pro­gram in a ­public Uni­ver­sity Hos­pital (PUH) in the ­periphery of ­Greece, and a pri­vate insti­tu­tion (PI) ­with an expe­ri­enced inten­sive ­care ­unit (ICU) in the cap­ital ­city of ­Athens. Pre­op­er­a­tive ­risk strat­ifi­ca­tion, mor­tality and post­op­er­a­tive ­length of ­stay (LOS) ­were ana­lysed ­between 1999 to 2001.
­Results. At PUH 298 ­patients ­were oper­ated and 312 ­patients at PI. ­There ­were 136 low ­risk (EuroS­CORE 0-2) and 474 ­medium and ­high-­risk ­patients (EuroS­CORE ≥3). ­There was no sig­nif­i­cantly ele­vated mor­tality or ­learning ­curve in low ­risk sur­gery ­either at PUH (57 ­patients ­with 1 ­death) or PI (79 ­patients 1 ­death). In ­medium and ­high-­risk sur­gery at PI ­there was no mor­tality in 68 ­patients oper­ated by the ­senior sur­geon and no ­learning ­curve in all 233 ­such ­patients. In 240 ­medium and ­high-­risk ­patients at PUH ­there was a ­learning ­curve ­despite the involve­ment of the ­same ­senior sur­geon. In 1999 and 2000 the ­observed mor­tality (OM) in 150 ­patients was 15.33%, EuroS­CORE 5.98, and in 2001 in 91 ­patients OM 3.29%, EuroS­CORE 5.95 ­with p=0.0038 ­when “expe­ri­enced” ICU ­staff was ­employed. LOS was sig­nif­i­cantly ­reduced in 97 ­patients in 2001 at PUH (8.7 d ± 2.81 vs 11.07­days ± 7.9 in 1999 and 2000, p=0.046) con­firming the exis­tence of a ­learning ­curve at the PUH. No ­such ­change was ­observed at PI (8.2 ­days vs 7.8, p=0.45).
Con­clu­sion. No mor­tality dif­fer­ences or ­learning ­curve char­ac­ter­is­tics ­were ­detected for low ­risk oper­a­tions ­either at PUH or PI. For ­medium and ­high ­risk sur­gery ­there ­appears to be a ­learning ­curve in PUH but not in PI ­despite ­senior sur­geon involve­ment in ­both. The pres­ence of an expe­ri­enced ICU ­appears to ­play a crit­ical ­role in the out­come of oper­a­tions in ­newly ­opened car­di­oth­o­racic pro­grams.

language: English


top of page