- Email: [email protected]
Total joint arthroplasty and incidence of postoperative bacteriuria with an indwelling catheter or intermittent catheterization with one-dose antibiotic prophylaxis
The Journal of Arthroplasty Vol. 16 No. 7 2001
Total Joint Arthroplasty and Incidence of Postoperative Bacteriuria With an Indwelling Catheter or Intermittent Catheterization With One-Dose Antibiotic Prophylaxis A Prospective Randomized Trial I. C. J. B. van den Brand, MD, and R. M. Castelein, MD, PhD
Abstract: This study examined the difference in postoperative bacteriuria in total joint arthroplasty after use of either an indwelling catheter or intermittent catheterization. Previous studies showed a preference for an indwelling catheter over intermittent catheterization to resolve postoperative urinary retention in total joint arthroplasty, but these studies generally used 48 hours of antibiotic prophylaxis. Increasing awareness of costs and bacterial resistance to antibiotics have prompted many centers to reduce prophylaxis to only 1 preoperative dose A prospective, randomized, controlled trial was conducted in primary total hip and primary total knee arthroplasty patients. One dose of cefazolin, 1 g, was administered intravenously immediately preoperatively. Five of 13 (38%) men in the indwelling catheter group and 0 of 14 (0%) men in the intermittent catheterization group developed postoperative bacteriuria (P ⫽ .016), and 6 of 33 (18%) women in the indwelling catheter group and 3 of 39 (8%) women in the intermittent catheterization group developed postoperative bacteriuria (not significant). A total of 11 (24%) patients in the indwelling catheter group (n ⫽ 46) and 3 (6%) patients in the intermittent catheterization group (n ⫽ 53) developed postoperative bacteriuria (P ⫽ .018). In this setting with 1-dose antibiotic prophylaxis, intermittent catheterization resulted in a lower incidence of postoperative bacteriuria compared with an indwelling catheter. For men, this difference is significant. Key words: antibiotic prophylaxis, indwelling catheter, intermittent catheterization, hip arthroplasty, knee arthroplasty, postoperative bacteriuria.
Urinary retention is a common postoperative problem in total hip arthroplasty (THA) and total knee
arthroplasty (TKA). Postoperative urinary retention often necessitates an indwelling Foley catheter or intermittent catheterization [1,2]. Urinary retention itself as well as catheterization can lead to bacteriuria [2– 4]. The chance of a deep infection in THA and TKA is increased 3 to 6 times if postoperative bacteriuria is present [5– 8]. Several case reports document postoperative hematogenous seeding from the urinary tract [3,9 –11]. Previous studies comparing an indwelling catheter and intermittent catheterization in THA and TKA favored an indwelling catheter because it led
From the Department of Orthopaedic Surgery, Isala Clinics, location Weezenlanden, Zwolle, The Netherlands. Submitted December 20, 2000; accepted April 12, 2001. No benefits or funds were received in support of this study. Reprint requests: R. M. Castelein, MD, PhD, Department of Orthopaedic Surgery, Isala Clinics, location Weezenlanden, PO Box 10500, 8000 GM Zwolle, The Netherlands. E-mail: [email protected] Copyright © 2001 by Churchill Livingstone威 0883-5403/01/1607-0006$35.00/0 doi:10.1054/arth.2001.25547
850
Postoperative Bacteriuria in THA and TKA • van den Brand and Castelein
851
Table 1. Eligible Patients, Exclusions, Enrollment, Withdrawals, and Number of Patients Who Completed the Study Eligible Exclusions due to criteria (24%) Enrollment Withdrawals (12%)
Completed study
148 3 29 3 113 3 2 7 2 99
to less bacteriuria and less urinary retention [12– 16]. These studies had different perioperative antibiotic prophylaxis protocols, however, varying in the antibiotics administered as well as the duration of administration, although all antibiotics were administered for at least 48 hours. Regarding the most appropriate perioperative regimen of antibiotic prophylaxis in total joint surgery, debate still exists [17,18]. There is a tendency to decrease the number of doses [19 –21]. Generally, antibiotic prophylaxis is aimed at per-operative bacteremia, but postoperative bacteremia also may threaten the survival of the prosthesis. Currently, as a result of increased cost awareness and the problem of bacterial resistance, in many hospitals only a single dose of antibiotic prophylaxis is administered intravenously immediately preoperatively to THA and TKA patients [19,20]. The aim of our study was to determine whether an indwelling catheter or intermittent catheterization leads to less postoperative bacteriuria or an urinary tract infection with 1 dose of cefazolin prophylaxis in primary THA or TKA.
Patients and Methods A prospective randomized trial at our hospital was conducted between April 1997 and May 1998. The inclusion criteria were patients undergoing primary THA or TKA. Informed consent was obtained, and patients were given the opportunity to refuse or withdraw. Patients with a history of chronic or recurrent urinary tract infections, preoperative or perioperative steroid medication, long-term antibiotic therapy, endocarditis antibiotic prophylaxis, or preoperative bacteriuria or urinary tract infection were excluded from this study (Table 1). Sex, age, weight, operation procedure, type of anesthetics, and concomitant diseases, such as diabetes mellitus and rheumatoid arthritis, were noted. All enrolled patients were randomized into 2 groups according to their patient number, which is
(29 men, 119 women) (medical history) (preoperative bacteriuria, all female) (extra antibiotics) (28 men, 85 women) (Foley catheter after operation inserted) (Foley catheter removed after day 2) (no urine culture or sediment postoperative) (complications) (27 men, 72 women)
independent of age, sex, procedure, or medical history. Patients in the indwelling catheter group received an indwelling Foley catheter in the operating room just before the start of surgery. The Foley catheter was inserted under sterile conditions, connected to a closed-drainage system, and remained in place for 48 hours. Patients in the intermittent catheterization group were catheterized intermittently by a trained staff nurse every 6 hours, or earlier if the patient had the urge to void but was unable to, until spontaneous bladder emptying occured. All patients routinely received 1 dose of cefazolin, 1 g, intravenously at the induction of anesthesia, 15 minutes before skin incision or tourniquet inflation, to reach adequate cefazolin concentrations for soft tissue and bone [22]. No postoperative antibiotics were used. Urinalysis on a midstream clean-catch urine specimen was performed on the day before surgery and on the 2nd and 5th postoperative days. On the day before surgery, urinalysis consisted of a sediment to ensure the absence of bacteria or white blood cells preoperatively. After surgery, urinalysis consisted of a sediment on the 2nd day and a sediment and culture on the 5th day. For the patients in the indwelling catheter group, the urine specimen on the 2nd day was taken after removal of the catheter. If a positive sediment for bacteria or white blood cells was present on the 2nd day, urine collection for culture was done the same day, and proper oral antibiotic treatment was started afterwards. Postoperative bacteriuria or a urinary tract infection was defined as a positive urine sediment for bacteria or white blood cells with a positive urine culture of ⬎ 100,000 colonies. All the results of the urinalysis preoperatively and postoperatively were reviewed independently by urologist blinded for the patient data. For the intermittent catheterization group, it was noted how often it was necessary to catheterize before spontaneous bladder emptying
852 The Journal of Arthroplasty Vol. 16 No. 7 October 2001 occurred. No prophylactic catheterization was performed immediately postoperatively because of a greater risk for bacteriuria [3,13,23]. For statistical analysis, the 2-tailed Fisher’s exact test and the exact test for linear trend in a cross-table were used.
Of the 148 patients eligible for this study, 29 (20%) were excluded because of a preoperative positive urinalysis. These 29 patients with asymptomatic bacteriuria all were older women (29 of 119 eligible women). These 29 women were operated (outside the study) with additional proper antibiotics because a preoperative asymptomatic bacteriuria has not been proved to be a risk factor for deep infection in THA and TKA when treated properly [24,25]. None of the men enrolled had a preoperative positive urinalysis. A total of 99 patients, 72 women and 27 men, completed the study after exclusions and withdrawals (Table 1). Of these patients, 46 were assigned randomly to the indwelling catheter group and 53 to the intermittent catheterization group. There were no significant differences in sex, age, operation, anesthetic procedure, or rheumatoid arthritis incidence between the study groups (Table 2). Nine diabetes patients were eligible, of whom 5 completed the study. These 5 patients all were assigned to group 1 by chance. Of the other 4 diabetes patients, 2 were excluded because of a positive preoperative urinalysis, and 2 were withdrawn because of a urine culture or sediment not obtained postoperatively. There were 14 patients (5 men, 9 women) with a postoperative bacteriuria, which is an overall inci-
Table 2. Distribution of Sex, Age, Operation, and the Prevalence of Diabetes and Rheumatoid Arthritis
THA TKA Spinal anesthesia General anesthesia Diabetes Rheumatoid arthritis
Men Women Total
Results
No. patients Men Women Mean age ⫹SD (range)
Table 3. Postoperative Bacteriuria
Indwelling Catheter
Intermittent Catheterization
46 13 33 68.6 ⫾ 8.8 (42–85) 26 20 25 21 5 5
53 14 39 68.2 ⫾ 9.0 (36–84) 35 18 25 28 0 5
THA, total hip arthroplasty; TKA, total knee arthroplasty.
Indwelling Catheter
Intermittent Catheterization
Overall
5/13 ⫽ 38%* 6/33 ⫽ 18%†
0/14 ⫽ 0%* 3/39 ⫽ 8%†
5/27 ⫽ 19% 9/72 ⫽ 13%
11/46 ⫽ 24%‡
3/53 ⫽ 6%‡
14/99 ⫽ 14%
*P ⫽ .016 (95% confidence interval, 12– 65). †P ⫽ .28 (not significant). ‡P ⫽ .018 (95% confidence interval, 4 –32).
dence of 14%. Of these 14 patients, 2 had a positive urinalysis and culture on the 2nd postoperative day; the other 12 patients had a positive result on the 5th postoperative day. The men in the indwelling catheter group had a bacteriuria rate of 38% (5 of 13 men) compared with 0% (0 of 14 men) in the intermittent catheterization group (P ⫽ .016). The women in the indwelling catheter group had a bacteriuria rate of 18% (6 of 33 women) compared with 8% (3 of 39 women) in the intermittent catheterization group (not significant). Overall, figures showed that the indwelling catheter group had a bacteriuria rate of 24% (11 of 46 patients) compared with 6% (3 of 53 patients) in the intermittent catheterization group (P ⫽ .018) (Table 3). In the indwelling catheter group, no patient needed straight catheterization after removal of the Foley catheter. For the intermittent catheterization group, the average number of catheterizations was 1.6 per patient (83 times in 53 patients). Fifteen (28%) of the 53 patients, 9 women (23%) and 6 men (43%), in the intermittent catheterization group needed no straight catheterization at all. None of these 15 noncatheterized patients developed a postoperative bacteriuria. The patients that were catheterized intermittently but did not develop a bacteriuria had an average of 1.9 catheterizations (67 times in 35 patients), and the 3 patients in this group who did develop a postoperative bacteriuria had an average of 5.3 straight catheterizations. The patients in the intermittent catheterization group that underwent general anesthesia were catheterized intermittently at an average of 1.6 times per patient compared with 1.5 times per patient among patients who underwent spinal anesthesia. For patients in this group who did not need any catheterization, general anesthesia was slightly more favorable, 32% (9 of the 28 patients), compared with spinal anesthesia, 24% (6 of the 25 patients). Comparison between general and spinal anesthesia and its influence on postoperative bac-
Postoperative Bacteriuria in THA and TKA • van den Brand and Castelein
853
Table 4. Number of Intermittent Catheterizations and Relationship to Postoperative Bacteriuria* Catheterizations
No. patients Bacteriuria
0
1
2
3
4
5
6
7
8
Total
15 0
17 0
10 1
7 0
1 0
0 0
1 1
1 0
1 1
53 3
*P ⫽ .003 (95% confidence interval, 10 –100).
teriuria revealed that 6 of the 49 patients (12%) with general anesthesia and 8 of the 50 patients (16%) with spinal anesthesia developed a postoperative bacteriuria. For statistical analysis, the catheterized patients in the intermittent catheterization group were divided into groups for number of catheterizations and the relationship to postoperative bacteriuria. This relationship between the number of catheterizations and postoperative bacteriuria was significant (P ⫽.003 Table 4). Of the 5 diabetes patients in the indwelling group, only 1 patient developed postoperative bacteriuria. Of the 10 patients with rheumatoid arthritis, equally divided between both groups, 2 patients (1 from each group) developed a postoperative bacteriuria. The postoperative positive cultures are listed in Table 5. Escherichia coli accounted for 60% of all positive cultures. In 1 patient, E. coli and Enterococcus faecalis were cultured. There were 2 minor complications, both in the intermittent catheterization group. The first patient developed slight hematuria during intermittent catheterization and received an indwelling Foley catheter. With the second patient, there was difficulty sliding in the catheter, and because of this it was left in place for 24 hours. These 2 patients could not complete this study and were withdrawn (Table 1). Neither of the 2 patients developed a bacteriuria or other urologic problems postoperatively.
Table 5. Postoperative Positive Cultures
Escherichia coli Enterococcus faecalis Klebsiella pneumoniae Acinetobacter wolffii -hemolytic streptococcus Proteus mirabilis
Indwelling Catheter
Intermittent Catheterization
Total
7 1 2 1 1 0
2 0 0 0 0 1
9 1 2 1 1 1
NOTE. One patient in group 1 had Escherichia coli and Enterococcus faecalis in 1 culture.
The costs for material and nursing time to insert a urinary catheter in our hospital were $6.15 (material $4.15 and nursing time $2) for an indwelling catheter and $4.85 (material $1.85 and nursing time $3) for each intermittent catheterization. The costs for nursing time were calculated as follows: with a mean monthly income of $1900, each 5 minutes costs about $1 per nurse. For inserting an indwelling catheter, 2 nurses are occupied for 5 minutes ($2), and for each intermittent catheterization, 1 nurse is occupied for 15 minutes ($3) [16]. With an average of 1.6 catheterizations per intermittent catheterized patient, this is a little more expensive than an indwelling catheter ($7.75 vs $6.15). The extra 5 minutes of occupying the operation room while inserting an indwelling catheter is not taken into account.
Discussion Postoperative bacteriuria after joint arthroplasty poses a threat to the survival of the prosthesis [5– 8]. Previous studies favored the use of an indwelling catheter over intermittent catheterization to treat urinary retention after THA or TKA because it caused less bacteriuria, although this difference was not significant [12–16]. In these studies, antibiotic prophylaxis was given for at least 48 hours postoperatively. Currently, because of increased cost awareness and the problem of bacterial resistance, in many hospitals only 1 dose of antibiotic prophylaxis is administered intravenously immediately preoperatively to THA or TKA patients [19,20]. We believed that this change in prophylaxis could have an effect on the postoperative incidence of bacteriuria after an indwelling catheter or intermittent catheterization. In this study of primary THA and TKA patients with 1 preoperative dose of cefazolin, we found a significantly higher percentage of postoperative bacteriuria after an indwelling catheter for 48 hours compared with intermittent catheterization (24% vs 6%; P ⫽ .018). Possible explanations for this difference are as follows.
854 The Journal of Arthroplasty Vol. 16 No. 7 October 2001 Our patients were not under antibiotic protection during the presence of the indwelling catheter compared with other studies, in which patients were under antibiotic protection for this period. In the other studies, perioperative antibiotic prophylaxis for at least 48 hours was administered. The various antibiotic prophylaxis protocols included cefamandole for 72 hours postoperatively by Lampe et al [14], cefazolin for 48 hours perioperatively and postoperatively by Knight and Pellegrini [16] and Oishi et al [15], cefazolin or clindamycin for 72 hours perioperatively and postoperatively by Carpiniello et al [13] and a cephalosporin with or without gentamicin perioperatively according to a standard protocol by Michelson et al [12]. The postoperative incidence of bacteriuria or urinary tract infection in the indwelling catheter groups of these studies ranged from 4% to 28% (28%, 8%, 4%, 11% and 4%). The postoperative incidence in our indwelling catheter group was 24%, which is comparatively high. Postoperative pain management was conducted without epidural analgesics. The number of intermittent catheterizations in our group was 1.6 compared with 8 times per patient in the study by Knight and Pellegrini [16], in which 82% of the patients received postoperative epidural analgesics for pain management. A postoperative incidence of bacteriuria of 12% occurred in their intermittent catheterization group. Epidural analgesics have a negative influence on bladder function [25–27], and perioperative catheterization causes an increased risk of bacteriuria or urinary tract infection [1,3,4,11,23,28,29]. Intermittent catheterization was not performed directly after surgery in the recovery room. Hozack et al [23] and Carpiniello et al [13] showed that in female patients the incidence of a positive postoperative urine culture was 16% if the intermittent catheterization was performed directly after surgery in the recovery room and 8.7% if no intermittent catheterization was performed in the recovery room. This 8.7% incidence resembles the 8% incidence of bacteriuria we found in female patients in the intermittent catheterization group. According to the protocol, the patients in the intermittent catheterization group were catheterized only if no spontaneous bladder emptying occurred in the first 6 hours after operation. Of the patients in the intermittent catheterization group, 28% (15 of 53) did not need any catheterization at all. None of these noncatheterized patients developed a bacteriuria (Table 4). Intermittent catheterization has been reported to produce an infection rate of 0.5% to 20% per catheterization [4,30]. In
our study, we found a significantly increased risk of postoperative bacteriuria in patients with more intermittent catheterizations (Table 4). For the men in this study, we found a high incidence of postoperative bacteriuria in the indwelling catheter group; compared with the men in the intermittent catheterization group, this difference is striking (38% vs 0%). Of the men in the intermittent catheterization group, 43% (6 of 14) did not need any catheterization at all. Redfern et al [31] found that 79% of the men in their study did not need postoperative catheterization. The main difference is that Redfern et al [31] performed ondemand catheterization instead of catheterization every 6 hours postoperatively in the absence of spontaneous bladder emptying. The type of anesthesia was not related to developing postoperative bacteriuria, which is in agreement with other studies [12,14,26]. There was no relationship between type of anesthesia and times of intermittent catheterization. The postoperative positive urine cultures showed mainly E. coli (60%). The presence of E. coli was relatively high compared with postoperative positive urine cultures obtained after a longer duration of antibiotic prophylaxis (39% by Glynn and Sheehan [24], 35% by Lampe et al. [14], 25% by Knight and Pellegrini [16]). This difference may be explained by the fact that the patients in our study did not receive antibiotics postoperatively; this resulted in low selection pressure on the bacterial colonization of the patients. The small extra cost ($1.60) of intermittent catheterization may not be a reason to insert an indwelling catheter in patients. Because time in the operating room is scarce and expensive, extra time in the operating room needed to insert an indwelling catheter is a disadvantage.
Conclusion The results of our study indicate that in the setting of primary THA and TKA with only 1 dose of cefazolin prophylaxis, intermittent catheterization for urinary retention causes less bacteriuria compared with an indwelling catheter for 48 hours. This difference is significant for men. For women, the incidence is twice as much for an indwelling catheter, but this difference is not significant in our study. Under these operating conditions with only 1 dose of cefazolin preoperatively, we recommend intermittent catheterization to treat postoperative urinary retention.
Postoperative Bacteriuria in THA and TKA • van den Brand and Castelein
References 1. Petersen MS, Collins DN, Selakovich WG, Finkbeiner AE: Postoperative urinary retention associated with total hip and total knee arthroplasties. Clin Orthop 269:102, 1991 2. Skelly JM, Guyatt GH, Kalbfleisch R, et al: Management of urinary retention after surgical repair of hip fracture. Can Med Assoc J 146:1185, 1992 3. Donovan TL, Gordon RO, Nagel DA: Urinary infections in total hip arthroplasty. J Bone Joint Surg Am 58:1134, 1976 4. Schaeffer AJ: Catheter-associated bacteriuria. Urol Clin North Am 13:735, 1986 5. Fitzgerald RH Jr, Nolan DR, Ilstrup DM, et al: Deep wound sepsis following total hip arthroplasty. J Bone Joint Surg Am 59:847, 1977 6. Surin VV, Sundholm K, Backman L: Infection after total hip replacement: With special reference to a discharge from the wound. J Bone Joint Surg Br 65:412, 1983 7. Ritter MA, Fechtmann RW: Urinary tract sequela: possible influence on joint infections following total joint replacements. Orthopaedics 10:467, 1987 8. Wymenga AB, van Horn JR, Theeuwes A, et al: Perioperative factors associated with septic arthritis after arthroplasty, prospective multicenter study of 362 knee and 2651 hip operations. Acta Orthop Scand 63:665, 1992 9. Irvine R, Johnson BL, Amstutz HC: The relationship of genitourinary tract procedures and deep sepsis after total hip replacements. Surg Gynecol Obstet 139:701, 1974 10. Benson MK, Hughes SP: Infections following total hip replacement in a general hospital without special orthopaedic facilities. Acta Orthop Scand 46:968, 1975 11. Wroblewski BM, DelSel HJ: Urethral instrumentation and deep sepsis in total joint replacement. Clin Orthop 146:209, 1980 12. Michelson JD, Lotke PA, Steinberg ME: Urinary bladder management after total joint replacement surgery. N Engl J Med 319:321, 1988 13. Carpiniello VL, Malloy TR, Cendron M, et al: Treatment of urinary complications after total joint replacement in elderly females. Urology 33:186, 1988 14. Lampe HIH, Sneller ZW, Rijnberg WJ: Problems with micturition after total hip replacement: to give or not to give an indwelling catheter. Ned Tijdschr Geneeskd 136:827, 1992 15. Oishi CS, Williams VJ, Hanson PB, et al: Periopera-
16. 17. 18. 19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30. 31.
855
tive bladder management after primary total hip arthroplasty. J Arthroplasty 10:732, 1995 Knight RM, Pellegrini VD: Bladder management after total joint arthroplasty. J Arthroplasty 11:882, 1996 Norden CW: Antibiotic prophylaxis in orthopedic surgery. Rev Infect Dis 13:S842, 1991 Schurman DJ, Woolson ST: Prophylactic antibiotics in orthopedic surgery. Orthopedics 7:1603, 1984 Wymenga AB: Joint sepsis after prophylaxis with one or three doses of cefuroxim in hip and knee replacement surgery. Thesis, University of Nijmegen, Nijmegen, The Netherlands, 1991 Mauerhan DR, Nelson CL, Smith DL, et al: Prophylaxis against infection in total joint arthroplasty: one day of cefuroxime compared with three days of cefazolin. J Bone Joint Surg Am 76:39, 1994 Hanssen AD, Osmon DR: The use of prophylactic antimicrobial agents during and after hip arthroplasty. Clin Orthop 369:124, 1999 Friedman RJ, Friedrich LV, White RL, et al: Antibiotic prophylaxis and tourniquet inflation in total knee arthroplasty. Clin Orthop 260:17, 1990 Hozack WJ, Carpiniello VL, Booth RE: The effect of early bladder catheterization on the incidence of urinary complications after total joint replacement. Clin Orthop 231:79, 1988 Glynn MK, Sheehan JM: The significance of asymptomatic bacteriuria in patients undergoing hip/knee arthroplasty. Clin Orthop 185:151, 1984 Husted S, Djurhuus JC, Husegaard HC, et al: Effect of postoperative extradural morphine on lower urinary tract function. Acta Anaesthesiol Scand 29:183, 1985 Walts LF, Kaufman RD, Moreland JR, et al: Total hip arthroplasty: an investigation of factors related to postoperative urinary retention. Clin Orthop 194: 280, 1985 Williams A, Price N, Willett K: Epidural anaesthesia and urinary dysfunction: the risks in total hip replacement. J R Soc Med 88:699P, 1995 Stamm WE, Hooton TM: Management of urinary tract infections in adults. N Engl J Med 329:1328, 1993 Stamm WE: Catheter associated urinary tract infections: epidemiology, pathogenesis, and prevention. Am J Med 91(suppl 3B):65S, 1991 Kunin CM: Genitourinary infections in the patient at risk: extrinsic risk factors. Am J Med 76:131, 1984 Redfern TR, Machin DG, Parsons KF, Owen R: Urinary retention in men after total hip arthroplasty. J Bone Joint Surg Am 68:1435, 1986
Total Joint Arthroplasty and Incidence of Postoperative Bacteriuria With an Indwelling Catheter or Intermittent Catheterization With One-Dose Antibiotic Prophylaxis A Prospective Randomized Trial I. C. J. B. van den Brand, MD, and R. M. Castelein, MD, PhD
Abstract: This study examined the difference in postoperative bacteriuria in total joint arthroplasty after use of either an indwelling catheter or intermittent catheterization. Previous studies showed a preference for an indwelling catheter over intermittent catheterization to resolve postoperative urinary retention in total joint arthroplasty, but these studies generally used 48 hours of antibiotic prophylaxis. Increasing awareness of costs and bacterial resistance to antibiotics have prompted many centers to reduce prophylaxis to only 1 preoperative dose A prospective, randomized, controlled trial was conducted in primary total hip and primary total knee arthroplasty patients. One dose of cefazolin, 1 g, was administered intravenously immediately preoperatively. Five of 13 (38%) men in the indwelling catheter group and 0 of 14 (0%) men in the intermittent catheterization group developed postoperative bacteriuria (P ⫽ .016), and 6 of 33 (18%) women in the indwelling catheter group and 3 of 39 (8%) women in the intermittent catheterization group developed postoperative bacteriuria (not significant). A total of 11 (24%) patients in the indwelling catheter group (n ⫽ 46) and 3 (6%) patients in the intermittent catheterization group (n ⫽ 53) developed postoperative bacteriuria (P ⫽ .018). In this setting with 1-dose antibiotic prophylaxis, intermittent catheterization resulted in a lower incidence of postoperative bacteriuria compared with an indwelling catheter. For men, this difference is significant. Key words: antibiotic prophylaxis, indwelling catheter, intermittent catheterization, hip arthroplasty, knee arthroplasty, postoperative bacteriuria.
Urinary retention is a common postoperative problem in total hip arthroplasty (THA) and total knee
arthroplasty (TKA). Postoperative urinary retention often necessitates an indwelling Foley catheter or intermittent catheterization [1,2]. Urinary retention itself as well as catheterization can lead to bacteriuria [2– 4]. The chance of a deep infection in THA and TKA is increased 3 to 6 times if postoperative bacteriuria is present [5– 8]. Several case reports document postoperative hematogenous seeding from the urinary tract [3,9 –11]. Previous studies comparing an indwelling catheter and intermittent catheterization in THA and TKA favored an indwelling catheter because it led
From the Department of Orthopaedic Surgery, Isala Clinics, location Weezenlanden, Zwolle, The Netherlands. Submitted December 20, 2000; accepted April 12, 2001. No benefits or funds were received in support of this study. Reprint requests: R. M. Castelein, MD, PhD, Department of Orthopaedic Surgery, Isala Clinics, location Weezenlanden, PO Box 10500, 8000 GM Zwolle, The Netherlands. E-mail: [email protected] Copyright © 2001 by Churchill Livingstone威 0883-5403/01/1607-0006$35.00/0 doi:10.1054/arth.2001.25547
850
Postoperative Bacteriuria in THA and TKA • van den Brand and Castelein
851
Table 1. Eligible Patients, Exclusions, Enrollment, Withdrawals, and Number of Patients Who Completed the Study Eligible Exclusions due to criteria (24%) Enrollment Withdrawals (12%)
Completed study
148 3 29 3 113 3 2 7 2 99
to less bacteriuria and less urinary retention [12– 16]. These studies had different perioperative antibiotic prophylaxis protocols, however, varying in the antibiotics administered as well as the duration of administration, although all antibiotics were administered for at least 48 hours. Regarding the most appropriate perioperative regimen of antibiotic prophylaxis in total joint surgery, debate still exists [17,18]. There is a tendency to decrease the number of doses [19 –21]. Generally, antibiotic prophylaxis is aimed at per-operative bacteremia, but postoperative bacteremia also may threaten the survival of the prosthesis. Currently, as a result of increased cost awareness and the problem of bacterial resistance, in many hospitals only a single dose of antibiotic prophylaxis is administered intravenously immediately preoperatively to THA and TKA patients [19,20]. The aim of our study was to determine whether an indwelling catheter or intermittent catheterization leads to less postoperative bacteriuria or an urinary tract infection with 1 dose of cefazolin prophylaxis in primary THA or TKA.
Patients and Methods A prospective randomized trial at our hospital was conducted between April 1997 and May 1998. The inclusion criteria were patients undergoing primary THA or TKA. Informed consent was obtained, and patients were given the opportunity to refuse or withdraw. Patients with a history of chronic or recurrent urinary tract infections, preoperative or perioperative steroid medication, long-term antibiotic therapy, endocarditis antibiotic prophylaxis, or preoperative bacteriuria or urinary tract infection were excluded from this study (Table 1). Sex, age, weight, operation procedure, type of anesthetics, and concomitant diseases, such as diabetes mellitus and rheumatoid arthritis, were noted. All enrolled patients were randomized into 2 groups according to their patient number, which is
(29 men, 119 women) (medical history) (preoperative bacteriuria, all female) (extra antibiotics) (28 men, 85 women) (Foley catheter after operation inserted) (Foley catheter removed after day 2) (no urine culture or sediment postoperative) (complications) (27 men, 72 women)
independent of age, sex, procedure, or medical history. Patients in the indwelling catheter group received an indwelling Foley catheter in the operating room just before the start of surgery. The Foley catheter was inserted under sterile conditions, connected to a closed-drainage system, and remained in place for 48 hours. Patients in the intermittent catheterization group were catheterized intermittently by a trained staff nurse every 6 hours, or earlier if the patient had the urge to void but was unable to, until spontaneous bladder emptying occured. All patients routinely received 1 dose of cefazolin, 1 g, intravenously at the induction of anesthesia, 15 minutes before skin incision or tourniquet inflation, to reach adequate cefazolin concentrations for soft tissue and bone [22]. No postoperative antibiotics were used. Urinalysis on a midstream clean-catch urine specimen was performed on the day before surgery and on the 2nd and 5th postoperative days. On the day before surgery, urinalysis consisted of a sediment to ensure the absence of bacteria or white blood cells preoperatively. After surgery, urinalysis consisted of a sediment on the 2nd day and a sediment and culture on the 5th day. For the patients in the indwelling catheter group, the urine specimen on the 2nd day was taken after removal of the catheter. If a positive sediment for bacteria or white blood cells was present on the 2nd day, urine collection for culture was done the same day, and proper oral antibiotic treatment was started afterwards. Postoperative bacteriuria or a urinary tract infection was defined as a positive urine sediment for bacteria or white blood cells with a positive urine culture of ⬎ 100,000 colonies. All the results of the urinalysis preoperatively and postoperatively were reviewed independently by urologist blinded for the patient data. For the intermittent catheterization group, it was noted how often it was necessary to catheterize before spontaneous bladder emptying
852 The Journal of Arthroplasty Vol. 16 No. 7 October 2001 occurred. No prophylactic catheterization was performed immediately postoperatively because of a greater risk for bacteriuria [3,13,23]. For statistical analysis, the 2-tailed Fisher’s exact test and the exact test for linear trend in a cross-table were used.
Of the 148 patients eligible for this study, 29 (20%) were excluded because of a preoperative positive urinalysis. These 29 patients with asymptomatic bacteriuria all were older women (29 of 119 eligible women). These 29 women were operated (outside the study) with additional proper antibiotics because a preoperative asymptomatic bacteriuria has not been proved to be a risk factor for deep infection in THA and TKA when treated properly [24,25]. None of the men enrolled had a preoperative positive urinalysis. A total of 99 patients, 72 women and 27 men, completed the study after exclusions and withdrawals (Table 1). Of these patients, 46 were assigned randomly to the indwelling catheter group and 53 to the intermittent catheterization group. There were no significant differences in sex, age, operation, anesthetic procedure, or rheumatoid arthritis incidence between the study groups (Table 2). Nine diabetes patients were eligible, of whom 5 completed the study. These 5 patients all were assigned to group 1 by chance. Of the other 4 diabetes patients, 2 were excluded because of a positive preoperative urinalysis, and 2 were withdrawn because of a urine culture or sediment not obtained postoperatively. There were 14 patients (5 men, 9 women) with a postoperative bacteriuria, which is an overall inci-
Table 2. Distribution of Sex, Age, Operation, and the Prevalence of Diabetes and Rheumatoid Arthritis
THA TKA Spinal anesthesia General anesthesia Diabetes Rheumatoid arthritis
Men Women Total
Results
No. patients Men Women Mean age ⫹SD (range)
Table 3. Postoperative Bacteriuria
Indwelling Catheter
Intermittent Catheterization
46 13 33 68.6 ⫾ 8.8 (42–85) 26 20 25 21 5 5
53 14 39 68.2 ⫾ 9.0 (36–84) 35 18 25 28 0 5
THA, total hip arthroplasty; TKA, total knee arthroplasty.
Indwelling Catheter
Intermittent Catheterization
Overall
5/13 ⫽ 38%* 6/33 ⫽ 18%†
0/14 ⫽ 0%* 3/39 ⫽ 8%†
5/27 ⫽ 19% 9/72 ⫽ 13%
11/46 ⫽ 24%‡
3/53 ⫽ 6%‡
14/99 ⫽ 14%
*P ⫽ .016 (95% confidence interval, 12– 65). †P ⫽ .28 (not significant). ‡P ⫽ .018 (95% confidence interval, 4 –32).
dence of 14%. Of these 14 patients, 2 had a positive urinalysis and culture on the 2nd postoperative day; the other 12 patients had a positive result on the 5th postoperative day. The men in the indwelling catheter group had a bacteriuria rate of 38% (5 of 13 men) compared with 0% (0 of 14 men) in the intermittent catheterization group (P ⫽ .016). The women in the indwelling catheter group had a bacteriuria rate of 18% (6 of 33 women) compared with 8% (3 of 39 women) in the intermittent catheterization group (not significant). Overall, figures showed that the indwelling catheter group had a bacteriuria rate of 24% (11 of 46 patients) compared with 6% (3 of 53 patients) in the intermittent catheterization group (P ⫽ .018) (Table 3). In the indwelling catheter group, no patient needed straight catheterization after removal of the Foley catheter. For the intermittent catheterization group, the average number of catheterizations was 1.6 per patient (83 times in 53 patients). Fifteen (28%) of the 53 patients, 9 women (23%) and 6 men (43%), in the intermittent catheterization group needed no straight catheterization at all. None of these 15 noncatheterized patients developed a postoperative bacteriuria. The patients that were catheterized intermittently but did not develop a bacteriuria had an average of 1.9 catheterizations (67 times in 35 patients), and the 3 patients in this group who did develop a postoperative bacteriuria had an average of 5.3 straight catheterizations. The patients in the intermittent catheterization group that underwent general anesthesia were catheterized intermittently at an average of 1.6 times per patient compared with 1.5 times per patient among patients who underwent spinal anesthesia. For patients in this group who did not need any catheterization, general anesthesia was slightly more favorable, 32% (9 of the 28 patients), compared with spinal anesthesia, 24% (6 of the 25 patients). Comparison between general and spinal anesthesia and its influence on postoperative bac-
Postoperative Bacteriuria in THA and TKA • van den Brand and Castelein
853
Table 4. Number of Intermittent Catheterizations and Relationship to Postoperative Bacteriuria* Catheterizations
No. patients Bacteriuria
0
1
2
3
4
5
6
7
8
Total
15 0
17 0
10 1
7 0
1 0
0 0
1 1
1 0
1 1
53 3
*P ⫽ .003 (95% confidence interval, 10 –100).
teriuria revealed that 6 of the 49 patients (12%) with general anesthesia and 8 of the 50 patients (16%) with spinal anesthesia developed a postoperative bacteriuria. For statistical analysis, the catheterized patients in the intermittent catheterization group were divided into groups for number of catheterizations and the relationship to postoperative bacteriuria. This relationship between the number of catheterizations and postoperative bacteriuria was significant (P ⫽.003 Table 4). Of the 5 diabetes patients in the indwelling group, only 1 patient developed postoperative bacteriuria. Of the 10 patients with rheumatoid arthritis, equally divided between both groups, 2 patients (1 from each group) developed a postoperative bacteriuria. The postoperative positive cultures are listed in Table 5. Escherichia coli accounted for 60% of all positive cultures. In 1 patient, E. coli and Enterococcus faecalis were cultured. There were 2 minor complications, both in the intermittent catheterization group. The first patient developed slight hematuria during intermittent catheterization and received an indwelling Foley catheter. With the second patient, there was difficulty sliding in the catheter, and because of this it was left in place for 24 hours. These 2 patients could not complete this study and were withdrawn (Table 1). Neither of the 2 patients developed a bacteriuria or other urologic problems postoperatively.
Table 5. Postoperative Positive Cultures
Escherichia coli Enterococcus faecalis Klebsiella pneumoniae Acinetobacter wolffii -hemolytic streptococcus Proteus mirabilis
Indwelling Catheter
Intermittent Catheterization
Total
7 1 2 1 1 0
2 0 0 0 0 1
9 1 2 1 1 1
NOTE. One patient in group 1 had Escherichia coli and Enterococcus faecalis in 1 culture.
The costs for material and nursing time to insert a urinary catheter in our hospital were $6.15 (material $4.15 and nursing time $2) for an indwelling catheter and $4.85 (material $1.85 and nursing time $3) for each intermittent catheterization. The costs for nursing time were calculated as follows: with a mean monthly income of $1900, each 5 minutes costs about $1 per nurse. For inserting an indwelling catheter, 2 nurses are occupied for 5 minutes ($2), and for each intermittent catheterization, 1 nurse is occupied for 15 minutes ($3) [16]. With an average of 1.6 catheterizations per intermittent catheterized patient, this is a little more expensive than an indwelling catheter ($7.75 vs $6.15). The extra 5 minutes of occupying the operation room while inserting an indwelling catheter is not taken into account.
Discussion Postoperative bacteriuria after joint arthroplasty poses a threat to the survival of the prosthesis [5– 8]. Previous studies favored the use of an indwelling catheter over intermittent catheterization to treat urinary retention after THA or TKA because it caused less bacteriuria, although this difference was not significant [12–16]. In these studies, antibiotic prophylaxis was given for at least 48 hours postoperatively. Currently, because of increased cost awareness and the problem of bacterial resistance, in many hospitals only 1 dose of antibiotic prophylaxis is administered intravenously immediately preoperatively to THA or TKA patients [19,20]. We believed that this change in prophylaxis could have an effect on the postoperative incidence of bacteriuria after an indwelling catheter or intermittent catheterization. In this study of primary THA and TKA patients with 1 preoperative dose of cefazolin, we found a significantly higher percentage of postoperative bacteriuria after an indwelling catheter for 48 hours compared with intermittent catheterization (24% vs 6%; P ⫽ .018). Possible explanations for this difference are as follows.
854 The Journal of Arthroplasty Vol. 16 No. 7 October 2001 Our patients were not under antibiotic protection during the presence of the indwelling catheter compared with other studies, in which patients were under antibiotic protection for this period. In the other studies, perioperative antibiotic prophylaxis for at least 48 hours was administered. The various antibiotic prophylaxis protocols included cefamandole for 72 hours postoperatively by Lampe et al [14], cefazolin for 48 hours perioperatively and postoperatively by Knight and Pellegrini [16] and Oishi et al [15], cefazolin or clindamycin for 72 hours perioperatively and postoperatively by Carpiniello et al [13] and a cephalosporin with or without gentamicin perioperatively according to a standard protocol by Michelson et al [12]. The postoperative incidence of bacteriuria or urinary tract infection in the indwelling catheter groups of these studies ranged from 4% to 28% (28%, 8%, 4%, 11% and 4%). The postoperative incidence in our indwelling catheter group was 24%, which is comparatively high. Postoperative pain management was conducted without epidural analgesics. The number of intermittent catheterizations in our group was 1.6 compared with 8 times per patient in the study by Knight and Pellegrini [16], in which 82% of the patients received postoperative epidural analgesics for pain management. A postoperative incidence of bacteriuria of 12% occurred in their intermittent catheterization group. Epidural analgesics have a negative influence on bladder function [25–27], and perioperative catheterization causes an increased risk of bacteriuria or urinary tract infection [1,3,4,11,23,28,29]. Intermittent catheterization was not performed directly after surgery in the recovery room. Hozack et al [23] and Carpiniello et al [13] showed that in female patients the incidence of a positive postoperative urine culture was 16% if the intermittent catheterization was performed directly after surgery in the recovery room and 8.7% if no intermittent catheterization was performed in the recovery room. This 8.7% incidence resembles the 8% incidence of bacteriuria we found in female patients in the intermittent catheterization group. According to the protocol, the patients in the intermittent catheterization group were catheterized only if no spontaneous bladder emptying occurred in the first 6 hours after operation. Of the patients in the intermittent catheterization group, 28% (15 of 53) did not need any catheterization at all. None of these noncatheterized patients developed a bacteriuria (Table 4). Intermittent catheterization has been reported to produce an infection rate of 0.5% to 20% per catheterization [4,30]. In
our study, we found a significantly increased risk of postoperative bacteriuria in patients with more intermittent catheterizations (Table 4). For the men in this study, we found a high incidence of postoperative bacteriuria in the indwelling catheter group; compared with the men in the intermittent catheterization group, this difference is striking (38% vs 0%). Of the men in the intermittent catheterization group, 43% (6 of 14) did not need any catheterization at all. Redfern et al [31] found that 79% of the men in their study did not need postoperative catheterization. The main difference is that Redfern et al [31] performed ondemand catheterization instead of catheterization every 6 hours postoperatively in the absence of spontaneous bladder emptying. The type of anesthesia was not related to developing postoperative bacteriuria, which is in agreement with other studies [12,14,26]. There was no relationship between type of anesthesia and times of intermittent catheterization. The postoperative positive urine cultures showed mainly E. coli (60%). The presence of E. coli was relatively high compared with postoperative positive urine cultures obtained after a longer duration of antibiotic prophylaxis (39% by Glynn and Sheehan [24], 35% by Lampe et al. [14], 25% by Knight and Pellegrini [16]). This difference may be explained by the fact that the patients in our study did not receive antibiotics postoperatively; this resulted in low selection pressure on the bacterial colonization of the patients. The small extra cost ($1.60) of intermittent catheterization may not be a reason to insert an indwelling catheter in patients. Because time in the operating room is scarce and expensive, extra time in the operating room needed to insert an indwelling catheter is a disadvantage.
Conclusion The results of our study indicate that in the setting of primary THA and TKA with only 1 dose of cefazolin prophylaxis, intermittent catheterization for urinary retention causes less bacteriuria compared with an indwelling catheter for 48 hours. This difference is significant for men. For women, the incidence is twice as much for an indwelling catheter, but this difference is not significant in our study. Under these operating conditions with only 1 dose of cefazolin preoperatively, we recommend intermittent catheterization to treat postoperative urinary retention.
Postoperative Bacteriuria in THA and TKA • van den Brand and Castelein
References 1. Petersen MS, Collins DN, Selakovich WG, Finkbeiner AE: Postoperative urinary retention associated with total hip and total knee arthroplasties. Clin Orthop 269:102, 1991 2. Skelly JM, Guyatt GH, Kalbfleisch R, et al: Management of urinary retention after surgical repair of hip fracture. Can Med Assoc J 146:1185, 1992 3. Donovan TL, Gordon RO, Nagel DA: Urinary infections in total hip arthroplasty. J Bone Joint Surg Am 58:1134, 1976 4. Schaeffer AJ: Catheter-associated bacteriuria. Urol Clin North Am 13:735, 1986 5. Fitzgerald RH Jr, Nolan DR, Ilstrup DM, et al: Deep wound sepsis following total hip arthroplasty. J Bone Joint Surg Am 59:847, 1977 6. Surin VV, Sundholm K, Backman L: Infection after total hip replacement: With special reference to a discharge from the wound. J Bone Joint Surg Br 65:412, 1983 7. Ritter MA, Fechtmann RW: Urinary tract sequela: possible influence on joint infections following total joint replacements. Orthopaedics 10:467, 1987 8. Wymenga AB, van Horn JR, Theeuwes A, et al: Perioperative factors associated with septic arthritis after arthroplasty, prospective multicenter study of 362 knee and 2651 hip operations. Acta Orthop Scand 63:665, 1992 9. Irvine R, Johnson BL, Amstutz HC: The relationship of genitourinary tract procedures and deep sepsis after total hip replacements. Surg Gynecol Obstet 139:701, 1974 10. Benson MK, Hughes SP: Infections following total hip replacement in a general hospital without special orthopaedic facilities. Acta Orthop Scand 46:968, 1975 11. Wroblewski BM, DelSel HJ: Urethral instrumentation and deep sepsis in total joint replacement. Clin Orthop 146:209, 1980 12. Michelson JD, Lotke PA, Steinberg ME: Urinary bladder management after total joint replacement surgery. N Engl J Med 319:321, 1988 13. Carpiniello VL, Malloy TR, Cendron M, et al: Treatment of urinary complications after total joint replacement in elderly females. Urology 33:186, 1988 14. Lampe HIH, Sneller ZW, Rijnberg WJ: Problems with micturition after total hip replacement: to give or not to give an indwelling catheter. Ned Tijdschr Geneeskd 136:827, 1992 15. Oishi CS, Williams VJ, Hanson PB, et al: Periopera-
16. 17. 18. 19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30. 31.
855
tive bladder management after primary total hip arthroplasty. J Arthroplasty 10:732, 1995 Knight RM, Pellegrini VD: Bladder management after total joint arthroplasty. J Arthroplasty 11:882, 1996 Norden CW: Antibiotic prophylaxis in orthopedic surgery. Rev Infect Dis 13:S842, 1991 Schurman DJ, Woolson ST: Prophylactic antibiotics in orthopedic surgery. Orthopedics 7:1603, 1984 Wymenga AB: Joint sepsis after prophylaxis with one or three doses of cefuroxim in hip and knee replacement surgery. Thesis, University of Nijmegen, Nijmegen, The Netherlands, 1991 Mauerhan DR, Nelson CL, Smith DL, et al: Prophylaxis against infection in total joint arthroplasty: one day of cefuroxime compared with three days of cefazolin. J Bone Joint Surg Am 76:39, 1994 Hanssen AD, Osmon DR: The use of prophylactic antimicrobial agents during and after hip arthroplasty. Clin Orthop 369:124, 1999 Friedman RJ, Friedrich LV, White RL, et al: Antibiotic prophylaxis and tourniquet inflation in total knee arthroplasty. Clin Orthop 260:17, 1990 Hozack WJ, Carpiniello VL, Booth RE: The effect of early bladder catheterization on the incidence of urinary complications after total joint replacement. Clin Orthop 231:79, 1988 Glynn MK, Sheehan JM: The significance of asymptomatic bacteriuria in patients undergoing hip/knee arthroplasty. Clin Orthop 185:151, 1984 Husted S, Djurhuus JC, Husegaard HC, et al: Effect of postoperative extradural morphine on lower urinary tract function. Acta Anaesthesiol Scand 29:183, 1985 Walts LF, Kaufman RD, Moreland JR, et al: Total hip arthroplasty: an investigation of factors related to postoperative urinary retention. Clin Orthop 194: 280, 1985 Williams A, Price N, Willett K: Epidural anaesthesia and urinary dysfunction: the risks in total hip replacement. J R Soc Med 88:699P, 1995 Stamm WE, Hooton TM: Management of urinary tract infections in adults. N Engl J Med 329:1328, 1993 Stamm WE: Catheter associated urinary tract infections: epidemiology, pathogenesis, and prevention. Am J Med 91(suppl 3B):65S, 1991 Kunin CM: Genitourinary infections in the patient at risk: extrinsic risk factors. Am J Med 76:131, 1984 Redfern TR, Machin DG, Parsons KF, Owen R: Urinary retention in men after total hip arthroplasty. J Bone Joint Surg Am 68:1435, 1986