Indwelling catheter increases the risk of urinary tract infection in total knee arthroplasty: A meta-analysis of randomized controlled trials.

BACKGROUND: The purpose of this meta-analysis was to assess whether a urinary catheter is necessary for all patients in primary total knee arthroplasty (TKA).
METHODS: PubMed, EMBASE, the Cochrane Library, Web of Science, and China National Knowledge Infrastructure were systematically searched for randomized controlled trials (RCTs). All RCTs were compared with receive either an indwelling urinary catheter or no urinary catheter in TKA. Primary outcomes were urinary retention and urinary tract infection. Secondary outcomes were the length of stay, duration of the surgery, and the first urination time.
RESULTS: A total of 6 RCTs involving 1334 patients were included in the meta-analysis. No significant difference between the 2 groups was found in urinary retention (P = .52), length of stay (P = .38), duration of the surgery (P = .55). However, patients with an indwelling catheter were associated with a higher risk of urinary tract infections and longer time for the first urination than patients without indwelling catheters (P = .009 and P = .004).
CONCLUSION: The available evidence indicates that patients without using the indwelling catheters could reduce urinary tract infections and the time for the first urination, without increase in the incidence of urinary retention in primary TKA. LEVEL OF EVIDENCE: Level I, therapeutic study.

Introduction

Postoperative urinary retention is a common complication after total knee arthroplasty (TKA).[ Indwelling or intermittent urinary catheterization is an effective treatment agent for prevention and treatment of postoperative urinary retention, and it is routinely used in TKA.[ However, indwelling catheters increases the risk factors for postoperative complications, such as urinary retention after removal of the catheter,[ damage to the urethral mucosa,[ and dysuria or urinary tract irritation.[ In addition, it is more important that urine retention may potentially increase urinary tract infections, which can lead to hematogenous bacteremia, and a risk of subsequent joint infection in TKA.[

In recent years, surgical techniques and fast-track guidelines have made great strides in TKA. The operation time of TKA is extremely shortened since the hemostasis technology is used to reduce the amount of blood loss, and advanced anesthesia technology is used to better monitor the vital signs during the operation. Some studies have reported the use of 500 to 600 mL of residual urine as an indicator of indwelling catheters, because excessive residual urine would cause excessive stretching of the bladder wall, resulting in long time difficulty in urination.[ A recent randomized controlled trial found that it is feasible to relax the threshold of bladder residual urine to 800 mL.[

Although many studies have been investigated, the urinary catheterization could be avoided in patients undergoing TKA[; however, to our knowledge, there is no conclusive evidence of the efficacy of whether a patient in TKA requires indwelling catheterization. Therefore, based on the current clinical studies with urinary catheterization, we aimed to pool the results of published randomized controlled trials (RCTs) to determine the efficacy of catheter management in TKA. The purpose of this meta-analysis was to investigate whether urinary catheterization was associated with: urinary retention, urinary tract infection, length of stay, duration of the surgery, and the first urination time.

Material and methods

This meta-analysis was based on the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions[ and was written in accordance with the PRISMA checklist (Preferred Reporting Items for Systematic Reviews and Meta-analyses). The meta-analysis extracts relevant data from published studies, so no ethics review approval was required.

Search strategy

PubMed (1996–October 2020), Embase (1980–October 2020), Web of Science (1990–October 2020) the Cochrane Library (October 2020), and the China National Knowledge Infrastructure (October 2020) were systematically searched. All RCTs were collected from related references to find additional studies in TKA. The keywords used were “total knee arthroplasty,” “total knee replacement,” “TKA,” “TKR,” “catheterize,” “indwelling catheter,” “catheterization” in conjunction with Boolean operators “AND” or “OR.” The 2 authors independently reviewed the title and abstract of the search to exclude significantly unrelated studies. The search steps of the flow diagram are presented in Fig. 1.

Figure 1

Flow diagram of literature selection.

Inclusion criteria and exclusion criteria

All studies were eligible in the meta-analysis if they met the following: participants: patients are primary TKA; intervention: patients receive an indwelling urinary in the control group and no urinary catheter in the study group. Outcomes: urinary retention, urinary tract infection, length of stay, duration of the surgery, and the first urination time; study design: all studies are RCTs. Studies that those without the outcome measures of interest were excluded. Letters, abstracts, comments, editorials, and studies without full text available were also excluded. The dispute is resolved through consultation, or the third reviewer is consulted.

Data extraction

Two authors (YPL, MYS) independently extracted the available data from the included studies. The data included first author, disease, publication date, patient, age, sex, intervention, the standard for urinary retention, and anesthesia. The primary outcome consisted of urinary retention and urinary tract infection. Secondary outcomes consisted of the length of stay, duration of the surgery, and the first urination time. All extracted data is summarized in a form in a pre-designed Microsoft Excel worksheet. Disagreements were resolved by discussion to receive consensus.

Quality assessment

Two authors (YPL, SMY) independently evaluated the risk of bias of the included studies according to the Cochrane Handbook, which consisted of the following items: “random sequence generation,” “allocation concealment,” “blinding of participants and personnel,” “blinding of outcomes assessment,” “incomplete outcome data,” “selective outcome reporting,” and “other bias.”[ The methodological quality of each included study was measured as “low bias,” “unclear bias,” and “high bias.”

Data analysis and statistical methods

The current meta-analysis was calculated using Review Manager Software Windows 5.3 (Cochrane Collaboration, Oxford, UK). For continuous outcomes, the mean difference (MD) with 95% confidence intervals (CIs) was used. For discontinuous outcomes, relative risk (RR) with 95% CIs was used. The statistical heterogeneity was assessed by the value of P and I2 using the chi-square test. If the I2 > 50% or P < .05 were considered to demonstrate significant heterogeneity and the random-effect model was chosen, otherwise the fixed-effect model was chosen. This meta-analysis also used a funnel plot of urinary tract infection to independently assess publication bias.

Results

Search results

A total of 193 studies were identified through the initial search, and 106 papers were excluded due to the duplicates. In the next stage, 87 papers were excluded after reading the title and abstract. Therefore, 6 RCTs[ were included after reading the full text according to the inclusion criteria: 241 patients in the control group and 255 patients in the study group.

Study characteristics

The baseline characteristics of the included studies are summarized in Table 1. All the studies were published between 2000 and 2018. The sample size ranged from 30 to 346, and the mean age ranged from 57.7 to 68.9 years. All the catheterization times reported in the studies were 24 hours after surgery, and the standard for urinary retention was >400 mL in most studies. All RCTs reported anesthetic methods such as spinal or epidural general,[ general anesthesia with saphenous nerve block,[ intravenous combined inhaled anesthesia,[ intravenous general anesthesia,[ spinal anesthesia,[ combined spinal, and epidural anesthesia.[

Table 1

Characteristics of included studies.

					Gender (M/F)		Intervention method
First author	Disease	Date	Patient (C/S)	Age (C/S)	C	S	C	S	Anesthesia	Urinary retention
Iorio[14]	n.s.	2000	306/346	67.8/66.8	143/163	162/184	24 h postoperative catheter	No	Spinal or epidural or general anesthesia	n.s.
Huang[15]	OA	2014	157/157	66.9/67.4	33/124	38/119	24 h postoperative catheter	No	General anesthesia with saphenous nerve block	400 mL
Cai[17]	OA/RA	2014	56/56	57.7/58.6	34/22	32/24	24 h postoperative catheter	No	Intraveous combined inhaled anesthesia	400 mL
Luo[18]	OA	2017	30/30	68.9/67.2	11/19	13/17	24 h postoperative catheter	No	Intravenous general anesthesia	400 mL
You[19]	OA	2017	50/50	62.7/64.2	22/28	24/26	24 h postoperative catheter	no	Spinal anesthesia	500 mL
Peng[20]	OA/RA	2018	48/48	66.8/67.4	n.s.	n.s.	24 h postoperative catheter	no	Combined spinal and epidural anesthesia	500 mL

Risk of bias in included studies

Table 2 and Fig. 2 summaries the risk of bias assessment of the 6 included studies. All the RCTs described the random sequence generation, allocation concealment, incomplete outcome data, selective reporting, and other biases. However, only one study[ reports the blinding method. The meta-analysis independently used urinary tract infection funnel plots to assess publication bias; the results were generally symmetrical and exhibited lower publication bias (Fig. 3).

Table 2

Quality assessment of included studies.

First author	Adequate randomization method	Allocation concealment	Blinding of participants and personnel	Blinding of the outcome assessment	Incomplete outcome	Free of selective reporting	Free of other bias	Level of evidence
Iorio[14]	Yes	Unclear	Unclear	Unclear	Yes	Yes	Yes	Level I
Huang[15]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Level I
Cai[17]	Yes	Yes	Unclear	Unclear	Yes	Yes	Yes	Level I
Luo[18]	Yes	Yes	Unclear	Unclear	Yes	Yes	Yes	Level I
You[19]	Yes	Yes	Unclear	Unclear	Yes	Yes	Yes	Level I
Peng[20]	Yes	Yes	Unclear	Unclear	Yes	Yes	Yes	Level I

Figure 2

Risk of bias summary.

Figure 3

Meta-analysis of funnel plot.

Meta-analysis results

Urinary retention

The urinary retention was reported in 4 studies[ with a total of 582 patients: 291 patients in the control group and 291 patients in the study group. The results of the meta-analysis showed no significant difference between the control group and study group (RR, 0.83; 95% CI, 0.46 to 1.48; P = .52). A fixed-effects model was applied due to the low statistical heterogeneity in the meta-analysis (P = .32, I2 = 15%) (Fig. 4).

Figure 4

Meta-analysis of urinary retention.

Urinary tract infection

The urinary tract infection was reported in all studies[ with a total of 1334 patients: 647 patients in the control group and 687 patients in the study group. The results of the meta-analysis showed a higher risk of urinary tract infection in the control group compared with the study group (RR, 2.72; 95% CI, 1.28–5.76; P = .009). A fixed-effects model was applied due to the low statistical heterogeneity in the meta-analysis (P = .38, I2 = 6%) (Fig. 5).

Figure 5

Meta-analysis of urinary tract infection.

Length of stay (days)

The length of stay was reported in 2 studies[ with a total of 426 patients: 213 patients in the control group and 213 patients in the study group. The results of the meta-analysis showed no significant difference between the control group and study group (MD, 0.15; 95% CI, –0.19–0.49; P = .38). A fixed-effects model was applied due to the low statistical heterogeneity in the meta-analysis (P = .53, I2 = 0%) (Fig. 6).

Figure 6

Meta-analysis of length of stay.

Duration of the surgery (minutes)

The duration of the surgery was reported in 4 studies[ with a total of 582 patients: 291 patients in the control group and 291 patients in the study group. The results of the meta-analysis showed no significant difference between the control group and study group (MD, –0.37; 95% CI, –1.57–0.84; P = .55). A fixed-effects model was applied due to the low statistical heterogeneity in the meta-analysis (P = .79, I2 = 0%) (Fig. 7).

Figure 7

Meta-analysis of duration of the surgery.

The first urination time (hour)

The first urination time was reported in 2 studies[ with a total of 160 patients: 80 patients in the control group and 80 patients in the study group. The results of the meta-analysis showed patients in the study group significantly reduced the first urination time by a mean of 2.03 hours compared with the control group (95% CI, 0.66–3.40; P = .004). A random-effects model was applied due to the high statistical heterogeneity in the meta-analysis (P = .0002, I2 = 93%) (Fig. 8).

Figure 8

Meta-analysis of the first urination time.

Discussion

This is the first meta-analysis that compared whether receive either an indwelling urinary catheter or no urinary catheter before the surgery in TKA. The pooled results indicate that, according to available evidence, patients with an indwelling catheter were associated with a higher risk of urinary tract infections and longer time for the first urination than patients without indwelling catheters. However, there was no significant difference between indwelling catheters versus non-indwelling catheters in the urinary retention, length of stay, and duration of the surgery following TKA.

Postoperative urinary retention is a common complication in total joint arthroplasty. However, the mechanism of the occurrence of urinary retention is complex and unclear, which may be affected by various factors such as anesthetic methods, types of analgesic drugs, mental and psychological and bladder detrusor weakness.[ In a prospective, randomized, controlled study, Knight et al[ found that the incidence of urinary retention in patients without indwelling catheter was 35% compared with 19% in patients with indwelling catheter. A potential explanation for this is that most of the patients in this trial received epidural anesthesia and even epidural analgesia within 48 hours after surgery. Epidural anesthesia has a longer time for parasympathetic block, and bladder detrusor is mainly dominated by parasympathetic nerves. Therefore, epidural anesthesia is more likely to cause postoperative urinary retention. Miller et al[ reported 200 patients undergoing total hip arthroplasty were randomized to treatment with or without insertion of an indwelling urinary catheter, they found that there was no significant difference between the 2 groups in urinary retention (9.7% vs 2.8%).

Although patients with general anesthesia still may result in urinary retention due to existing urinary reflex inhibition. However, the postoperative pain scores in patients with general anesthesia are lower than those in the spinal anesthesia. As a result, it takes less opioid and obtains earlier functional activity, which contributes early bladder function recovery.[ In addition, general anesthesia has less effect on urinary reflexes than epidural and spinal anesthesia, then the postoperative bladder urinary function recovery is also faster than other anesthesia methods.[ Cai et al[ performed an RCT, in which all patients underwent general anesthesia with intravenous inhalation; they found a postoperative urinary retention rate of 5.4% for patients without an indwelling urinary catheter compared with a postoperative urinary retention rate of 3.6% for patients with an indwelling urinary catheter. The incidence of urinary retention was not statistically significant between the 2 groups. In addition, the author further found that there were still patients who had urinary retention after the removal of the catheter needed to re-indwelling the catheter, suggesting that indwelling catheter before surgery is not enough to avoid the risk of urinary retention. The results of the current meta-analysis are consistent with previous reports, showing that TKA without an indwelling catheter is safe and feasible

The meta-analysis also demonstrated that urinary tract infections were more significantly higher in patients with indwelling catheters. As previously reported by Huang et al,[ 314 patients following TKA were randomized either to receive or not receive a urinary catheter before surgery, they reported a urinary tract infection rate of 0.6% in patients without an indwelling urinary catheter compared with a urinary tract infection rate of 5.1% in patients with an indwelling urinary catheter. In another study by Miller et al,[ they reported a urinary tract infection rate of 2.8% in 107 patients undergoing THA with a urinary catheter. Urinary tract infections are thought to be predisposing factors for joint prosthesis infections, especially in patients receiving permanent joint implants. In addition, previous studies have shown that urinary tract infections account for 40% of all hospital-acquired infections,[ resulting in a huge burden of medical expenses. It is considered that infections in patients with urinary tract disease may require longer hospital stays for additional examinations and treatments. Therefore, prevention of urinary tract infections will become an important measure to reduce joint infections and medical costs. This study demonstrates that reducing the use of catheters may potentially reduce the incidence of joint infection in TKA.

In addition, the results in the meta-analysis demonstrated that patients in the study group significantly reduced the first urination time by a mean of 2.03 hours compared with the control group, which can also potentially increase the incidence of urine retention in patients with indwelling catheters. No significant difference between the 2 groups was found in length of stay and duration of the surgery.

This meta-analysis has several advantages. First, this is the first meta-analysis to compare the management of catheterization or non-urinary catheterization in TKA. Second, we only included RCTs with strict statistical calculations, and the quality of the included studies was relatively high. Third, postoperative urinary retention and urinary tract infections were the primary outcome in the current meta-analysis, for the both of these results were a response to the main effects and complications of the catheter. It appears that the prevalence of urinary tract infections is relatively higher with the use of an indwelling catheter in patients in TKA. Based on the above conclusions, therefore, we can draw hypotheses that patients are screened for the high risk of preoperative urinary retention and further reduce the risk factors for urinary retention. Then, most patients with TKA can potentially avoid catheterization, which has important clinical implications for reducing unnecessary catheterization and related complications, particularly in the urinary tract and periprosthetic infections.

Our meta-analysis also had several potential limitations: patients were administered with different the standard for urinary retention and the time of catheterization, which may lead to large heterogeneity; the follow-up time in the study was limited, and therefore some adverse events may be underestimated; and only 6 RCTs were included, and more high-quality studies is needed to confirm the above conclusions in the future.

Conclusions

Taking our evidence-based data into consideration for TKA patients, our meta-analysis results demons treated that compared without indwelling catheters, patients with indwelling catheters increased the incidence of urinary tract infections and prolonged the time of first urination. There was no significant difference in the urinary retention, length of stay, and duration of the surgery.

Author contributions

Mingying Shuai contributed to draft writing and data analysis. Yueping Li designed the study and participated in the data analysis. All authors agree to be accountable for all aspects of the work.

Conceptualization: Yueping Li.

Data curation: Mingying Shuai, Yueping Li.

Investigation: Mingying Shuai.

Methodology: Mingying Shuai.

Writing – original draft: Mingying Shuai, Yueping Li.

Writing – review & editing: Mingying Shuai, Yueping Li.