The use of a urinary bladder catheter in the perioperative period for patie
nts undergoing total knee arthroplasty is controversial, In the current stu
dy, two bladder management protocols were studied. One group of patients ha
d an indwelling catheter inserted into the bladder before total knee arthro
plasty. The other group of patients was observed and treated for urinary re
tention as necessary. From 1993 to 1998, 652 patients undergoing primary, u
nilateral total knee arthroplasty were randomized by surgeon into two group
s: one group underwent preoperative insertion of an indwelling bladder cath
eter (306 patients), and one group (346 patients) had a catheter inserted p
ostoperatively as necessary. Sixty-six percent (229 of 346) of these patien
ts required catheterization (203 had indwelling catheters and 26 had interm
ittent straight catheters), A urinary tract infection developed in five pat
ients (1.6%) in whom a catheter was inserted preoperatively, A urinary trac
t infection developed in six patients (1.7%) in whom a catheter was inserte
d if necessary. Five of these urinary tract infections developed in patient
s with delayed indwelling bladder catheters. A urinary tract infection did
not develop in any patient in whom a straight catheter was inserted. There
was no significant difference in the length of stay in the hospital between
the two groups. The group in whom a catheter always was inserted generated
$491 greater cost for total knee arthroplasty than patients in whom a cath
eter was inserted if necessary.