Hospital volume affects outcome after total knee arthroplasty. (2014)

• Record Type: Journal Article
• Authors: Pamilo, Konsta J; Peltola, Mikko; Paloneva, Juha; Mäkelä, Keijo; Häkkinen, Unto; Remes, Ville
• Is Part Of: Acta orthopaedica. Volume 86:Number 1(2015)
• Page Start: 41
• Page End: 47
• Publication Date: 2014
• Subjects: Orthopedics -- Periodicals; 616.7005
• DOI: 10.3109/17453674.2014.977168
• Languages: English
• ISSNs: 1745-3674

Abstract

<abstract> <title> <x xml:space="preserve">Abstract</x> </title> <p> <bold>Background and purpose —</bold> The influence of hospital volume on the outcome of total knee joint replacement surgery is controversial. We evaluated nationwide data on the effect of hospital volume on length of stay, re-admission, revision, manipulation under anesthesia (MUA), and discharge disposition for total knee replacement (TKR) in Finland.</p> <p> <bold>Patients and methods —</bold> 59, 696 TKRs for primary osteoarthritis performed between 1998 and 2010 were identified from the Finnish Hospital Discharge Register and the Finnish Arthroplasty Register. Hospitals were classified into 4 groups according to the number of primary and revision knee arthroplasties performed on an annual basis throughout the study period: 1–99 (group 1), 100–249 (group 2), 250–449 (group 3), and ≥ 450 (group 4). The association between hospital procedure volume and length of stay (LOS), length of uninterrupted institutional care (LUIC), re-admissions, revisions, MUA, and discharge disposition were analyzed.</p> <p> <bold>Results —</bold> The greater the volume of the hospital, the shorter was the average LOS and LUIC. Smaller hospital volume was not unambiguously associated with increased revision, re-admission, or MUA rates. The smaller the annual hospital volume, the more often patients were discharged home.</p> <p> <bold>Interpretation —</bold> LOS and LUIC ought to be shortened in lower-volume hospitals. There is potential for a reduction in length of stay in extended institutional care facilities.</p> </abstract>