The Journal of Bone & Joint Surgery

The Journal of Bone & Joint Surgery, Volume 89, Issue 4

Scientific Articles | April 01, 2007

Timing of Tourniquet Release in Knee Arthroplasty: Meta-Analysis of Randomized, Controlled Trials

Krishna Reddi Boddu Siva Rama, MRCS¹; Sunil Apsingi, MS²; Sharmila Poovali, MD³; Anand Jetti, MRCOG⁴

¹ Department of Musculoskeletal Surgery, Imperial College, 7 East, Charing Cross Hospital, Fulham Palace Road, London W6 8RF, United Kingdom. E-mail address: r.rama@imperial.ac.uk
² Department of Orthopaedics, Luton and Dunstable Hospital NHS Trust, Lewsey Road, Luton LU4 0DZ, United Kingdom
³ Ipswich Hospital, Heath Road, Ipswich IP4 5PD, United Kingdom
⁴ Glasgow Royal Infirmary, 84 Castle Street, Glasgow G4 0SF, Scotland

View Disclosures and Other Information

Disclosure: The authors did not receive any outside funding or grants in support of their research for or preparation of this work. Neither they nor a member of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a member of their immediate families, are affiliated or associated.

A commentary is available with the electronic versions of this article, on our web site () and on our quarterly CD-ROM (call our subscription department, at 781-449-9780, to order the CD-ROM).

Investigation performed at Hammersmith Hospitals NHS Trust, London, United Kingdom

The Journal of Bone and Joint Surgery, Incorporated

J Bone Joint Surg Am, 2007 Apr 01;89(4):699-705. doi: 10.2106/JBJS.F.00497

5 Recommendations (Recommend) | 3 Comments | Saved by 3 Users Save Case

Article

Comments (0)

text A A A

Abstract

Background: Some surgeons release the tourniquet before closing the wound to secure hemostasis during knee arthroplasty. We examined whether early tourniquet release could reduce the perioperative blood loss and whether not releasing the tourniquet until after wound closure would increase the risk of early postoperative complications.

Methods:We searched electronic databases and reference lists of relevant articles, retrieved all of the published randomized controlled trials designed to address these issues, and performed a meta-analysis.

Results:Eleven studies involving a total of 872 patients and 893 primary knee arthroplasties were analyzed systematically. The studies showed considerable clinical and methodological diversity. Early release of the tourniquet increased the total measured blood loss (weighted mean difference = 228.7 mL; 95% confidence interval = 168.3 to 289.1; p < 0.00001). Early release also increased blood loss as calculated on the basis of the maximum decrease in hemoglobin concentration (weighted mean difference = 320.7 mL; 95% confidence interval = 204.3 to 437.1; p < 0.00001). The rate of reoperations due to postoperative complications was 3.1% (nine of 290) in the group with late tourniquet release compared with 0.3% (one of 290) in the group with early tourniquet release; the risk difference was 3% (95% confidence interval, 0.1% to 5%), which was significant (p = 0.04).

Conclusions: Early tourniquet release for hemostasis increases the blood loss associated with primary knee arthroplasty. However, tourniquet release after wound closure can increase the risk of early postoperative complications requiring another operation. Well-conducted large studies are needed to further explore the risk of early postoperative complications associated with late tourniquet release in knee arthroplasty.

Level of Evidence: Therapeutic Level II. See Instructions to Authors for a complete description of levels of evidence.

Figures in this Article

The advantage of using a pneumatic tourniquet during knee arthroplasty to reduce blood loss is highly debatable in light of available evidence^1-6. However, it is a common practice mainly because it improves visualization in the surgical field and achieves a bloodless field for cementing. Release of the tourniquet before closure of the wound has been advocated to secure hemostasis, to avoid the risk of not recognizing a major vascular injury, and for better evaluation of patellar tracking. The effect of the timing of the tourniquet release on blood loss and early postoperative complications has been investigated in many studies^2,7-18. However, variations in the methodologies of these studies, small sample sizes, conflicting results, and wide confidence intervals of the results have limited the clinical utility of the findings. A systematic identification, appraisal, and synthesis of all of these studies, with a potential for statistical aggregation, can strengthen the link between the best evidence and the optimal clinical practice¹⁹.

We examined the question of whether tourniquet release before wound closure reduces the perioperative blood loss associated with knee arthroplasty. We also attempted to determine whether not releasing the tourniquet until after wound closure increases the risk of early postoperative complications following knee arthroplasty.

Materials and Methods

Materials and Methods | Results | Discussion | Appendix | References

We conducted this meta-analysis according to the methodological guidelines outlined by the Cochrane Collaboration (Oxford, United Kingdom)²⁰, and we reported our findings according to the recommendations outlined in the Quality of Reporting of Meta-Analyses (QUOROM) statement²¹.

Searching

The electronic databases that we searched included the Cochrane Central Register of Controlled Trials (2006), Ovid Medline and PubMed Medline (1966 to 2005), Embase (1980 to 2005), TrialsCentral (), and Current Controlled Trials (). For the search in Medline and Embase we used the following search terms and Boolean operators: tourniquet AND (knee OR arthroplasty OR joint replacement OR joint prosthesis). We also searched the reference lists of the relevant articles for additional trials. We contacted authors and experts to ask them to identify any additional relevant trials. The search was not restricted to specific languages or years of publication.

Selection

The criteria for inclusion of an article were (1) a study population treated with primary knee arthroplasty under tourniquet control, (2) a study comparing release of the tourniquet before wound closure (early release) to secure hemostasis in one group of patients with tourniquet release after wound closure (late release) in the other group, (3) blood loss and/or postoperative complications as the principal outcomes, (4) a randomized controlled trial. Two authors (K.R.B.S.R. and S.A.) independently reviewed the titles and, if a title suggested any likelihood of the study meeting the inclusion criteria, the abstract was reviewed to determine if the study was eligible for retrieval. We then critically appraised the articles and identified those that were eligible for inclusion in our study. Any disagreements were resolved by discussion.

Validity Assessment

Two authors (K.R.B.S.R. and S.P.) independently assessed the methodological quality of each study on the basis of whether it included allocation concealment, blinding, a report of the loss of patients during the study, and any selective reporting of outcomes. We also used the Consolidated Standards of Reporting Trials (CONSORT) checklist²² of twenty-two items to assess the quality of reporting. Disagreements were resolved by discussion.

Data Abstraction

The relevant data were extracted from the articles independently by two authors (K.R.B.S.R. and S.A.), after which the data were compared and disagreements were resolved if necessary. The one German article¹⁶ was translated by a medical translator. We attempted to contact the authors of the articles^7,18 with missing data. The data obtained from each study included the flow of participants through each stage of the study; sample sizes; demographic data; criteria for patient selection; type of arthroplasty; type of anesthesia; method of thromboprophylaxis; tourniquet time; operative time; rate of lateral retinacular release; usage of wound drains; rehabilitation protocol; and procedure for measuring blood loss and outcome measurements such as "postoperative blood loss," "total measured blood loss," "calculated blood loss," prevalence of early complications (within the first six weeks after the operation), fall in hemoglobin concentration, number of blood transfusions, and amount of blood transfused. Postoperative blood loss was defined as the blood measured in the negative suction wound drains after the operation. Total measured blood loss was defined as the sum of the intraoperative and postoperative blood loss. Intraoperative blood loss was measured by weighing the swabs and measuring the blood in the suction drain during the operation. The calculated blood loss was estimated with methods, as described by Bourke and Smith²³ and by Gross²⁴, based on the maximum decrease in hemoglobin concentration normalized to the patient's weight and height. Deep vein thrombosis was considered to be present if it was confirmed by investigations. Wound complications, such as erythema, a tense wound, hematoma, cellulitis, oozing, and dehiscence, that did not require another operation were grouped as minor wound complications. Wound dehiscence, hematoma, and deep infection that required drainage and/or débridement and knee stiffness that required manipulation under anesthesia were grouped as complications that required a reoperation.

Statistical Analysis

If the standard deviation of a mean was not reported in a study, it was calculated indirectly from the 95% confidence interval or p value. When necessary, we calculated the mean and standard deviation from the median and range with use of the methods proposed by Hozo et al.²⁵. Where appropriate, the results of comparable groups of trials were pooled to calculate weighted mean differences and risk differences with 95% confidence intervals, with use of a fixed-effects meta-analysis. We chose risk differences rather than risk ratios in order to avoid excluding studies with zero events in both of the groups. The inverse variance method was used for continuous outcomes, and the Mantel-Haenszel method²⁶ was used for dichotomous outcomes. Heterogeneity between comparable trials was measured with use of a standard chi-square test (with a level of significance of p = 0.1) as well as the I² statistic for heterogeneity²⁷. An I² statistic value of >50% was considered to indicate substantial heterogeneity. The random-effects model described by DerSimonian and Laird was used for meta-analysis of the continuous data with substantial heterogeneity²⁸ in order to obtain meaningful confidence intervals of the effect. Alternatively, we investigated for heterogeneity due to clinical diversity and methodological diversity by performing subgroup and sensitivity analyses when appropriate. The pooled effect was considered significant if the p value was <0.05. Statistical analysis was performed with the RevMan 4.2.8 software package (Cochrane Collaboration, Oxford, United Kingdom).

Results

Materials and Methods | Results | Discussion | Appendix | References

Study Characteristics

Of 1322 articles retrieved by the search, thirteen^2,7-18 that assessed the effect of the timing of tourniquet release on blood loss and/or early complications associated with knee arthroplasty were identified. Eleven of these thirteen studies were randomized controlled trials and were included in the meta-analysis^8-18. Ten studies were publications in indexed journals, and one was a conference abstract¹⁸. The eleven studies involved a total of 872 patients and 893 knee arthroplasties. The tourniquet was released before wound closure for hemostasis during 458 arthroplasties and after wound closure during 435 arthroplasties. The sample sizes ranged from twenty-five to fifty for both the early-release and late-release groups.

Clinical Diversity

The studies showed considerable clinical diversity. All involved primary knee arthroplasties only. Two studies involved uncemented, hybrid, and cemented knee replacements^9,13; one involved uncemented and cemented knee replacements¹⁰; two included only uncemented replacements^12,18; and six involved only cemented knee replacements^8,11,14-17. One study involved unicondylar knee replacements in addition to total knee replacements¹³. Information about lateral retinacular release for correction of patellar maltracking was reported in only four studies. In two of them^13,14, the rate of lateral release was reported to be similar in the two groups, although the exact data were not provided. In the other two studies^11,12, lateral release was not performed. Chemical thromboprophylaxis was carried out with aspirin in one study¹⁴, with warfarin in one¹¹, and with low-molecular-weight heparin in five^{8,10,13,16,17}. There was no thromboprophylaxis in one study¹², and a specific method was not mentioned in two studies^9,18. In one study, one group of patients was treated with and one was treated without heparin thromboprophylaxis¹⁵. The suction drain was left in place for twenty-four hours in four studies^11-13,17, for forty-eight hours in two studies^8,10, and until the drainage was <50 mL in a twelve-hour period or until the third day (whichever came first) in two studies^9,16; the drain protocol was unclear in three studies^14,15,18.

View Large | Download Slide(.ppt)
Add to My JBJS

Fig. 1: Individual and pooled weighted mean differences (WMD) in the total measured blood loss and the calculated blood loss between knee arthroplasties performed with early tourniquet release and those performed with late tourniquet release. A fixed-effects model was used for total measured blood loss, and a random effects model was used for calculated blood loss. In the study by Lotke et al.¹⁴, in one group (Lotke 1) continuous passive motion was started three days after the operation whereas in another group (Lotke 2) it was started immediately in the recovery room. SD = standard deviation, CI = confidence interval, and df = degrees of freedom.

Methodological Diversity

The studies also showed diversity with regard to reporting and methodology. The CONSORT scores for the quality of reporting ranged from 7 to 17 (of 22) points (mean and standard deviation, 13.8 ± 3.2 points). The conference abstract¹⁸ was given the lowest score (7 points). There was good agreement between the reviewers with regard to their assessments of the quality of reporting (r = 0.80). The quality of the methodology varied among the studies as well. Randomization allocation was concealed adequately in only two studies^8,11, and only one study was blinded⁸. No patients were lost after randomization in any study. In most of the studies, selective reporting of some of the minor wound complications could not be excluded as these studies were primarily designed to identify the influence of the timing of the tourniquet release on the blood loss^11-18. The interrater agreement was excellent for all of these components of quality assessment (? = 1.0).

Outcome Analysis

Combining the results of the five studies^{10,12,13,15,17} in which the total blood loss had been measured properly, as defined earlier, showed significantly more total measured blood loss in the patients with early release of the tourniquet than in those with late tourniquet release (weighted mean difference = 228.7 mL; 95% confidence interval = 168.3 to 289.1; p < 0.00001) (Fig. 1 and Table I). However, pooling of the data for measured postoperative blood loss in eight studies^9-14,16,17 showed no significant difference between the two groups (p = 0.2) (Table I). In three studies^12,14,16, the total blood loss had been calculated with use of the methods described by Gross²⁴, which are based on the maximum decrease in hemoglobin concentration normalized to the patient's weight and height. Combining the results of those studies demonstrated that the calculated blood loss was significantly greater in the early-release group (weighted mean difference = 320.7 mL; p < 0.00001) (Fig. 1 and Table I).

View Larger

TABLE I Weighted Mean Differences in Various Continuous Parameters Between the Early and Late-Tourniquet-Release Groups

Parameter	Pooled Mean for Early-Tourniquet-Release Group	Pooled Mean for Late-Tourniquet-Release Group	Weighted Mean Difference (95% Confidence Interval)	P Value
Total measured blood loss (mL)	930.6 (n = 196)	740.3 (n = 186)	228.7 (168.3 to 289.1)	<0.00001
Calculated blood loss (mL)	1372.8 (n = 119)	1126.1 (n = 132)	320.7 (204.3 to 437.1)	<0.00001
Measured postoperative blood loss (mL)	713.1 (n = 338)	701.7 (n = 337)	44.5 (-23.2 to 112.2)	0.20
Decrease in hemoglobin concentration in all patients (g/L)	28.9 (n = 165)	29.5 (n = 160)	0.2 (-2.8 to 3.3)	0.90
Volume of blood transfused (mL)	700.3 (n = 127)	620.8 (n = 120)	91.8 (4.6 to 179.1)	0.04
Operative time (min)	87.9 (n = 286)	81.8 (n = 279)	7.0 (2.8 to 11.2)	0.001
Tourniquet time (min)	69.3 (n = 127)	80.2 (n = 127)	-11.9 (-21.19 to -2.66)	0.002

Add to My JBJS

TABLE I Weighted Mean Differences in Various Continuous Parameters Between the Early and Late-Tourniquet-Release Groups

Parameter	Pooled Mean for Early-Tourniquet-Release Group	Pooled Mean for Late-Tourniquet-Release Group	Weighted Mean Difference (95% Confidence Interval)	P Value
Total measured blood loss (mL)	930.6 (n = 196)	740.3 (n = 186)	228.7 (168.3 to 289.1)	<0.00001
Calculated blood loss (mL)	1372.8 (n = 119)	1126.1 (n = 132)	320.7 (204.3 to 437.1)	<0.00001
Measured postoperative blood loss (mL)	713.1 (n = 338)	701.7 (n = 337)	44.5 (-23.2 to 112.2)	0.20
Decrease in hemoglobin concentration in all patients (g/L)	28.9 (n = 165)	29.5 (n = 160)	0.2 (-2.8 to 3.3)	0.90
Volume of blood transfused (mL)	700.3 (n = 127)	620.8 (n = 120)	91.8 (4.6 to 179.1)	0.04
Operative time (min)	87.9 (n = 286)	81.8 (n = 279)	7.0 (2.8 to 11.2)	0.001
Tourniquet time (min)	69.3 (n = 127)	80.2 (n = 127)	-11.9 (-21.19 to -2.66)	0.002

There was no significant difference between the two groups with regard to the drop in hemoglobin levels on the second or third postoperative day^8,9,13,16 (p = 0.9) (Table I). One study¹² demonstrated changes in hemoglobin levels, hematocrit values, and red blood-cell counts at one week after the operation and showed a greater decrease in these parameters in the early-release group (all p values <0.01). However, in the majority of the studies in which hemoglobin levels were reported, the authors neither considered blood transfusions nor measured the total blood loss. Significantly more blood was transfused in the patients in the early-release group^10,11,17 (p = 0.04) (Table I). Blood transfusion was required for 28.6% of the patients with early tourniquet release compared with 21.1% of those with late release^8,9,13,16,18 (number of patients in the two groups, 192 compared with 185; relative risk = 1.5; 95% confidence interval = 0.9 to 2.5; p = 0.09; I² = 0.6%). However, the criteria for making the decision to perform a blood transfusion varied among the studies.

Seven studies included data on early postoperative complications^{8-11,13,16,17}. No systemic complications were reported. Pooling of the results showed that late tourniquet release increased the risk of regional complications (wound complications, symptomatic deep venous thrombosis, and knee stiffness requiring manipulation) significantly (risk difference = 7%; 95% confidence interval, 2 to 12; p = 0.006) (Table II). Analysis of the subsets demonstrated the rate of another operation to be 3.1% (nine of 290) in the late-release group compared with 0.3% (one of 290) in the early-release group (p = 0.04) (Table II). In the late-release group, there were two cases of wound dehiscence that were treated with débridement and resuturing, three wound hematomas treated with surgical drainage, and four cases of knee stiffness treated with manipulation. A deep infection developed in one of the three knees with a hematoma and was treated with a two-stage revision arthroplasty. The only second operation in the early-release group was a manipulation with the patient under anesthesia. The risk difference for minor wound complications was not significant (p = 0.12) (Table II).

View Larger

TABLE II Pooled Risk Differences in Rates of Early Postoperative Complications Between the Late and Early-Tourniquet-Release Groups*

Event	Pooled Incidence in Late-Tourniquet-Release Group	Pooled Incidence in Early-Tourniquet-Release Group	Risk Difference (95% Confidence Interval)	P Value
All regional complications	14.8% (43/290)	7.9% (23/290)	7% (2 to 12)	0.006
Complications that required another operation†	3.1% (9/290)	0.3% (1/290)	3% (0.1 to 5)	0.04
Minor wound complications‡	10% (29/290)	6.6% (19/290)	3% (-1 to 8)	0.12
Symptomatic deep vein thrombosis§	1.7% (5/290)	1.0% (3/290)	1% (-2 to 3)	0.61

The data from seven studies^{8-11,13,16,17} in which complications were reported were pooled, and fixed-effects estimates of risk differences were calculated. †Wound dehiscence, hematomas, and infections that required drainage and/or débridement and knee stiffness that required manipulation with the patient under anesthesia. ‡Erythema, tense wounds, hematoma, cellulitis, oozing, and dehiscence that did not require another operation. §Deep vein thrombosis was confirmed in all instances by phlebography, which was performed only in patients in whom the complication was suspected clinically.

Add to My JBJS

TABLE II Pooled Risk Differences in Rates of Early Postoperative Complications Between the Late and Early-Tourniquet-Release Groups*

Event	Pooled Incidence in Late-Tourniquet-Release Group	Pooled Incidence in Early-Tourniquet-Release Group	Risk Difference (95% Confidence Interval)	P Value
All regional complications	14.8% (43/290)	7.9% (23/290)	7% (2 to 12)	0.006
Complications that required another operation†	3.1% (9/290)	0.3% (1/290)	3% (0.1 to 5)	0.04
Minor wound complications‡	10% (29/290)	6.6% (19/290)	3% (-1 to 8)	0.12
Symptomatic deep vein thrombosis§	1.7% (5/290)	1.0% (3/290)	1% (-2 to 3)	0.61

The studies^9,13,16 did not reveal the true prevalence of deep vein thrombosis, as the investigations (phlebography in all studies) were performed only in patients in whom deep vein thrombosis was suspected clinically (Table II).

The authors of three studies^8,15,17 reported that the mean amount of knee flexion was not different between the two groups as assessed up to six weeks to two months after the operation. Barwell et al.⁸, in a study of forty-four patients in both the early and the late-release group, reported higher pain scores (p = 0.01) in the late-release group at four hours but no differences in the analgesic requirement over twenty-four hours or in the mean hospital stay. They also reported a greater mean time until a straight-leg raise could be performed in the late-release group (average, five days compared with 2.8 days in the early-release group; p < 0.01) and an increased risk of excessive swelling of the knee in that group (p = 0.05), although the objectivity of such measurements was unclear.

Relevant subgroup and sensitivity analyses of the clinical and methodological diversity of the studies did not demonstrate any effects that were different from those reported above (see Appendix).

Discussion

Materials and Methods | Results | Discussion | Appendix | References

This meta-analysis demonstrated that early release of the tourniquet to achieve hemostasis increases perioperative blood loss in association with primary knee arthroplasty. However, the risk of a complication requiring additional surgery was significantly increased when the tourniquet was left inflated until wound closure was complete.

Our study had several limitations. We could not perform a statistical test to exclude publication bias because of our small sample size. However, our comprehensive searching and duplicate selection process minimized such bias. The investigators performing the validity assessment and data extraction were not blinded with regard to the author, journal, institution, or results, as blinding was shown not to alter the results of a meta-analysis^29,30. We did not use methodological quality scores because of the lack of empirical evidence of the usefulness of such scores^31,32. Although the methodology of the studies was variable, we included in the analysis all trials that met our inclusion criteria, to minimize the bias against the trials with underreporting. Moreover, we dealt with the diversity by performing sensitivity and subgroup analyses.

We also identified two nonrandomized trials in which the effects of the timing of tourniquet release had been assessed^2,7. Page et al.⁷ reported a reduction in the total measured blood loss when they released the tourniquet before wound closure. However, diathermy was used only in the early-release group, which explains the reduction in the blood loss. In a study of sixteen knees treated with early tourniquet release and thirty-six treated with late release, Harvey et al.² found that early tourniquet release increased the total measured blood loss (mean difference, 448 mL; p < 0.01). They also performed phlebography for all patients and found no difference in the prevalence of deep vein thrombosis between the two groups (31% for both).

Measured blood loss after total knee arthroplasty is generally an underestimation and can be only 50% of the "true" blood loss, as calculated with several methods^23,24,33,34. Although some³³ have attributed the so-called hidden blood loss to hemolysis, investigators using labeled red cells^35,36 showed that it is entirely due to perioperative bleeding. This finding suggests that the available methods of measuring intraoperative blood loss are inefficient. In our meta-analysis, early tourniquet release increased both the calculated blood loss and the total measured blood loss but not the postoperative blood loss, suggesting that intraoperative blood loss might have contributed considerably to the increase in total blood loss associated with the early tourniquet release.

Reactive hyperemia and an increase in the fibrinolytic activity occur for a few minutes after tourniquet release, leading to excessive bleeding^37-41. A closed wound and apposed tissue layers with a firmly applied dressing can produce a local compressive effect and control this bleeding. Early tourniquet release provides a window of time for bleeding to occur before such an effect is achieved. Even though it allows visualization of the bleeding points, it is difficult to identify and control all such points. Some authors have reduced such intraoperative blood loss by using a fibrin spray or by injecting the periarticular tissues with solutions containing local anesthetic and epinephrine^42,43. Moreover, bleeding from the cut cancellous bone has been observed to be a major source of blood loss in knee arthroplasty⁴⁴. The use of cement has been claimed to control this bone bleeding to some extent⁴⁴. Although knee arthroplasties with cement were associated with less blood loss than those done without cement^9,10,13,44, our meta-analysis showed early release of the tourniquet to have the same effect on blood loss in both types of arthroplasty (see Appendix).

This meta-analysis demonstrated no significant difference in the drop in the hemoglobin level (weighted mean difference = 0.2 g/L; p = 0.9) between the two groups despite the difference in total measured blood loss (weighted mean difference = 228.7 mL; p < 0.00001). Variability in the blood transfusion rates and the criteria for transfusion among the studies explains these findings.

There is a risk that a major vascular injury will not be recognized if the tourniquet is left inflated throughout the operation. The prevalence of major arterial complications after total knee replacement is low (0.17% [seven of 4097 knee replacements]⁴⁵ or once in every 107 surgeon working years⁴⁶), but the prognosis is poor if such a complication develops, with death or amputation occurring in many cases^45-49. However, most reported cases have been thrombotic injuries in patients with advanced atherosclerosis. Identifying such pre-existing arterial insufficiency and avoiding tourniquet use in these patients have been strongly recommended⁴⁶. Nevertheless, direct sharp injury to the artery has also been described, although it is believed to occur less often and to be associated with a better prognosis than arterial thrombosis^46,47. It is theoretically possible that late tourniquet release increases the risk of a delay in the recognition of such incidents. Avoiding the use of a tourniquet in all knee arthroplasties to prevent such serious complications has been proposed and debated⁴⁶.

The risk of complications leading to another operation was seen to be increased when the tourniquet remained inflated until after wound closure. Although the risk difference cannot be generalized because of the considerable reporting variability in the literature, sensitivity analysis including only the studies^8,10 in which the outcomes had been predefined (avoiding selective reporting) emphasized the increased risk of a second operation with late tourniquet release (see Appendix). Despite the fact that there was less total blood loss in these knees, inadequate control of subcutaneous bleeding points may cause wound hematomas. Such hematomas may produce local ischemia that can lead to wound dehiscence and infection. Moreover, evaluation of patellar tracking while the tourniquet is inflated may demonstrate spurious lateral tracking as a result of increased lateral retinacular tension⁵⁰. This could lead to an unnecessary lateral release, with its potential complications such as hematoma and reduced blood supply to the wound edges. Studies have shown a 1.5 to ninefold increase in the rate of lateral release when patellar tracking was evaluated in the presence of an inflated tourniquet^51-54. The rate of lateral releases was not reported in the majority of the studies in our meta-analysis.

The inadequate reporting and methodology in many of the studies in this meta-analysis justify additional randomized trials to explore the effect of the timing of tourniquet release on the risk of postoperative complications. Such studies should include a large number of patients, adequate concealment of randomization allocation, blinded assessment, proper predefinition of the outcomes, and full reporting of the results.

Appendix

Materials and Methods | Results | Discussion | Appendix | References

Tables showing the results of sensitivity and subgroup analyses of the outcomes are available with the electronic versions of this article, on our web site at jbjs.org (go to the article citation and click on "Supplementary Material") and on our quarterly CD-ROM (call our subscription department, at 781-449-9780, to order the CD-ROM). ?

References

Materials and Methods | Results | Discussion | Appendix | References

Abdel-Salam A, Eyres KS. Effects of tourniquet during total knee arthroplasty. A prospective randomised study. J Bone Joint Surg Br. 1995;77: 250-3.77250 1995 [PubMed]

Harvey EJ, Leclerc J, Brooks CE, Burke DL. Effect of tourniquet use on blood loss and incidence of deep vein thrombosis in total knee arthroplasty. J Arthroplasty. 1997;12: 291-6.12291 1997 [PubMed][CrossRef]

Tetro AM, Rudan JF. The effects of a pneumatic tourniquet on blood loss in total knee arthroplasty. Can J Surg. 2001;44: 33-8.4433 2001

Vandenbussche E, Duranthon LD, Couturier M, Pidhorz L, Augereau B. The effect of tourniquet use in total knee arthroplasty. Int Orthop. 2002;26: 306-9.26306 2002 [PubMed][CrossRef]

Wakankar HM, Nicholl JE, Koka R, D'Arcy JC. The tourniquet in total knee arthroplasty. A prospective, randomised study. J Bone Joint Surg Br. 1999;81: 30-3.8130 1999 [PubMed][CrossRef]

Wauke K, Nagashima M, Kato N, Ogawa R, Yoshino S. Comparative study between thromboembolism and total knee arthroplasty with or without tourniquet in rheumatoid arthritis patients. Arch Orthop Trauma Surg. 2002;122: 442-6.122442 2002 [PubMed]

Page MH, Shepherd BD, Harrison JM. Reduction of blood loss in knee arthroplasty. Aust NZ J Surg. 1984;54: 141-4.54141 1984 [CrossRef]

Barwell J, Anderson G, Hassan A, Rawlings I. The effects of early tourniquet release during total knee arthroplasty: a prospective randomized double-blind study. J Bone Joint Surg Br. 1997;79: 265-8. Erratum in: J Bone Joint Surg Br. 1977;79:693.79265 1997 [CrossRef]

Burkart BC, Bourne RB, Rorabeck CH, Kirk PG, Nott L. The efficacy of tourniquet release in blood conservation after total knee arthroplasty. Clin Orthop Relat Res. 1994;299: 147-52.299147 1994 [PubMed]

Christodoulou AG, Ploumis AL, Terzidis IP, Chantzidis P, Metsovitis SR, Nikiforos DG. The role of timing of tourniquet release and cementing on perioperative blood loss in total knee replacement. Knee. 2004;11: 313-7.11313 2004 [PubMed][CrossRef]

Hersekli MA, Akpinar S, Ozkoc G, Ozalay M, Uysal M, Cesur N, Tandogan RN. The timing of tourniquet release and its influence on blood loss after total knee arthroplasty. Int Orthop. 2004;28: 138-41.28138 2004 [CrossRef]

Ishii Y, Matsuda Y. Effect of the timing of tourniquet release on perioperative blood loss associated with cementless total knee arthroplasty: a prospective randomized study. J Arthroplasty. 2005;20: 977-83.20977 2005 [CrossRef]

Jorn LP, Lindstrand A, Toksvig-Larsen S. Tourniquet release for hemostasis increases bleeding. A randomized study of 77 knee replacements. Acta Orthop Scand. 1999;70: 265-7.70265 1999 [PubMed][CrossRef]

Lotke PA, Faralli VJ, Orenstein EM, Ecker ML. Blood loss after total knee replacement. Effects of tourniquet release and continuous passive motion. J Bone Joint Surg Am. 1991;73: 1037-40.731037 1991 [PubMed]

Newman JH, Jackson JP, Waugh W. Timing of tourniquet removal after knee replacement. J R Soc Med. 1979;72: 492-4.72492 1979 [PubMed]

Schuh A, Hausel M, Salminen S. [Effect of tourniquet use on blood loss in total knee arthroplasty]. Zentralbl Chir. 2003;128: 866-70. German.128866 2003 [CrossRef]

Widman J, Isacson J. Surgical hemostasis after tourniquet release does not reduce blood loss in knee replacement. A prospective randomized study of 81 patients. Acta Orthop Scand. 1999;70: 268-70.70268 1999 [PubMed][CrossRef]

Yagi T, Ohno K, Ohtera K, Otokozawa C. The efficacy of tourniquet release in blood conservation following cementless total knee arthroplasty. Abstracts of 62nd annual meeting of American Academy of Orthopaedic Surgeons. 1995;19: 332.19332 1995

Gerbarg ZB, Horwitz RI. Resolving conflicting clinical trials: guidelines for meta-analysis. J Clin Epidemiol. 1988;41: 503-9.41503 1988 [CrossRef]

Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions 4.2.6. Updated 2006 Sep. . Accessed 2007 Jan 18.
http://www.cochrane.org/resources/handbook/Handbook4.2.6Sep2006.pdf

Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. Quality of Reporting of Meta-analyses. Lancet. 1999;354: 1896-900.3541896 1999 [PubMed][CrossRef]

Moher D, Schulz KF, Altman DG. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet. 2001;357: 1191-4.3571191 2001 [PubMed][CrossRef]

Bourke DL, Smith TC. Estimating allowable hemodilution. Anesthesiology. 1974;41: 609-12.41609 1974 [PubMed][CrossRef]

Gross JB. Estimating allowable blood loss: corrected for dilution. Anesthesiology. 1983;58: 277-80.58277 1983 [PubMed][CrossRef]

Hozo SP, Djulbegovic B, Hozo I. Estimating the mean and variance from the median, range, and the size of a sample. BMC Med Res Methodol. 2005;5: 13.513 2005 [PubMed][CrossRef]

Mantel N, Haenszel W. Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst. 1959;22: 719-48.22719 1959 [PubMed]

Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327: 557-60.327557 2003 [PubMed][CrossRef]

DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 1986;7: 177-88.7177 1986 [PubMed][CrossRef]

Berlin JA. Does blinding of readers affect the results of meta-analyses? University of Pennsylvania Meta-analysis Blinding Study Group. Lancet. 1997;350: 185-6.350185 1997 [PubMed][CrossRef]

Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, McQuay HJ. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials. 1996;17: 1-12.171 1996 [CrossRef]

Emerson JD, Burdick E, Hoaglin DC, Mosteller F, Chalmers TC. An empirical study of the possible relation of treatment differences to quality scores in controlled randomized clinical trials. Control Clin Trials. 1990;11: 339-52.11339 1990 [PubMed][CrossRef]

Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA. 1995;273: 408-12.273408 1995 [PubMed][CrossRef]

Sehat KR, Evans R, Newman JH. How much blood is really lost in total knee arthroplasty? Correct blood loss management should take hidden loss into account. Knee. 2000;7: 151-5.7151 2000 [PubMed][CrossRef]

Pattison E, Protheroe K, Pringle RM, Kennedy AC, Dick WC. Reduction in haemoglobin after knee joint surgery. Ann Rheum Dis. 1973;32: 582-4.32582 1973 [PubMed][CrossRef]

Erskine JG, Fraser C, Simpson R, Protheroe K, Walker ID. Blood loss with knee joint replacement. J R Coll Surg Edinb. 1981;26: 295-7.26295 1981

Mc Manus KT, Velchik MG, Alavi A, Lotke PA. Non-invasive assessment of postoperative bleeding in TKA patients with Tc-99m RBCs. Proceedings of the 34th annual meeting of the Society of Nuclear Medicine. June 2-5, 1987, Toronto, Ontario, Canada. Abstracts. J Nucl Med. 1987;28(4 Suppl): 565.28565 1987

Santavirta S, Hockerstedt K, Linden H. Pneumatic tourniquet and limb blood flow. Acta Orthop Scand. 1978;49: 565-70.49565 1978 [PubMed][CrossRef]

Silver R, de la Garza J, Rang M, Koreska J. Limb swelling after release of a tourniquet. Clin Orthop Relat Res. 1986;206: 86-9.20686 1986

Aglietti P, Baldini A, Vena LM, Abbate R, Fedi S, Falciani M. Effect of tourniquet use on activation of coagulation in total knee replacement. Clin Orthop Relat Res. 2000;371: 169-77.371169 2000 [PubMed][CrossRef]

Klenerman L, Chakrabarti R, Mackie I, Brozovic M, Stirling Y. Changes in haemostatic system after application of a tourniquet. Lancet. 1977;1: 970-2.1970 1977 [PubMed][CrossRef]

Fahmy NR, Patel DG. Hemostatic changes and postoperative deep-vein thrombosis associated with use of a pneumatic tourniquet. J Bone Joint Surg Am. 1981;63: 461-5.63461 1981 [PubMed]

Levy O, Martinowitz U, Oran A, Tauber C, Horoszowski H. The use of fibrin tissue adhesive to reduce blood loss and the need for blood transfusion after total knee arthroplasty. A prospective, randomized, multicenter study. J Bone Joint Surg Am. 1999;81: 1580-8.811580 1999 [PubMed]

Lombardi AV Jr, Berend KR, Mallory TH, Dodds KL, Adams JB. Soft tissue and intra-articular injection of bupivacaine, epinephrine, and morphine has a beneficial effect after total knee arthroplasty. Clin Orthop Relat Res. 2004;428: 125-30.428125 2004 [PubMed][CrossRef]

Mylod AG Jr, France MP, Muser DE, Parsons JR. Perioperative blood loss associated with total knee arthroplasty. A comparison of procedures performed with and without cementing. J Bone Joint Surg Am. 1990;72: 1010-2.721010 1990

Calligaro KD, Dougherty MJ, Ryan S, Booth RE. Acute arterial complications associated with total hip and knee arthroplasty. J Vasc Surg. 2003;38: 1170-7. Erratum in: J Vasc Surg. 2004;39:628.381170 2003 [PubMed][CrossRef]

Kumar SN, Chapman JA, Rawlins I. Vascular injuries in total knee arthroplasty. A review of the problem with special reference to the possible effects of the tourniquet. J Arthroplasty. 1998;13: 211-6.13211 1998 [CrossRef]

Da Silva MS, Sobel M; Surgeons of the Southern Association of Vascular Surgery. Popliteal vascular injury during total knee arthroplasty. J Surg Res. 2003;109: 170-4.109170 2003 [PubMed][CrossRef]

Holmberg A, Milbrink J, Bergqvist D. Arterial complications after knee arthroplasty: 4 cases and a review of the literature. Acta Orthop Scand. 1996;67: 75-8.6775 1996 [PubMed][CrossRef]

Rush JH, Vidovich JD, Johnson MA. Arterial complications of total knee replacement. The Australian experience. J Bone Joint Surg Br. 1987;69: 400-2.69400 1987 [PubMed]

Komatsu T, Ishibashi Y, Otsuka H, Nagao A, Toh S. The effect of surgical approaches and tourniquet application on patellofemoral tracking in total knee arthroplasty. J Arthroplasty. 2003;18: 308-12.18308 2003 [CrossRef]

Husted H, Toftgaard Jensen T. Influence of the pneumatic tourniquet on patella tracking in total knee arthroplasty: a prospective randomized study in 100 patients. J Arthroplasty. 2005;20: 694-7.20694 2005 [PubMed][CrossRef]

Marson BM, Tokish JT. The effect of a tourniquet on intraoperative patellofemoral tracking during total knee arthroplasty. J Arthroplasty. 1999;14: 197-9.14197 1999 [PubMed][CrossRef]

Lombardi AV Jr, Berend KR, Mallory TH, Dodds KL, Adams JB. The relationship of lateral release and tourniquet deflation in total knee arthroplasty. J Knee Surg. 2003;16: 209-14.16209 2003 [PubMed]

Laskin RS. Lateral release rates after total knee arthroplasty. Clin Orthop Relat Res. 2001;392: 88-93.39288 2001 [PubMed][CrossRef]

Topics

hemorrhage ; postoperative complications ; tourniquets ; wounds and injuries ; knee replacement arthroplasty

Username

Password

Access for Patients

Forgot Password?

Have a subscription to the print edition?

Not a Subscriber?

Buy this Article

Get online access for 30 days for $35

New to JBJS?

Subscribe to the online edition for 30 days for $75

Register for a FREE limited account to get full access to all CME activities, to comment on public articles, or to sign up for alerts.

Have a subscription to the print edition?

Current subscribers to The Journal of Bone & Joint Surgery in either the print or quarterly DVD formats receive free online access to JBJS.org.

Activate your online account now

Forgot your password?

Enter your username and email address. We'll send you a reminder to the email address on record.

Username:*

Email Address:*

Forgot your username or need assistance? Please contact customer service at subs@jbjs.org. If your access is provided
by your institution, please contact you librarian or administrator for username and password information. Institutional
administrators, to reset your institution's master username or password, please contact subs@jbjs.org

References

Accreditation Statement

These activities have been planned and implemented in accordance with the Essential Areas and policies of the Accreditation Council for Continuing Medical Education (ACCME) through the joint sponsorship of the American Academy of Orthopaedic Surgeons and The Journal of Bone and Joint Surgery, Inc. The American Academy of Orthopaedic Surgeons is accredited by the ACCME to provide continuing medical education for physicians.

CME Activities Associated with This Article

Submit a Comment

Please read the other comments before you post yours. Contributors must reveal any conflict of interest.
Comments are moderated and will appear on the site at the discretion of JBJS editorial staff.

* = Required Field

Comment Author(s) * (if multiple authors, separate names by comma)

Example: John Doe

Affiliation & Institution*

Comment Title*

Comment*

Cancel

Print
PDF
Email

Recipient(s) will receive an email with a link (good for 5 days) to 'Timing of Tourniquet Release in Knee Arthroplasty: Meta-Analysis of Randomized, Controlled Trials'.
Your Name:*
Example: John Doe

Email Address:* CC Me:
Enter your valid email address. Example: jdoe@example.com

Recipient's Email Address:*

Separate multiple email address with semi-colons (up to 5).

Subject:* 's The Journal of Bone & Joint Surgery: 'Timing of Tourniquet Release in Knee Arthroplasty: Meta-Analysis of Randomized, Controlled Trials'
Subject for your email.

Message:

(Optional, message will truncate at 1000 characters)

Processing your request... Please Wait...

Copyright © in the material you requested is held by The Journal of Bone & Joint Surgery, Inc. (unless otherwise noted). This email ability is provided as a courtesy, and by using it you agree that that you are requesting the material solely for personal, non-commercial use, and that it is subject to Journal of Bone & Joint Surgery, Inc.'s Terms of Use. The information provided in order to email this topic will not be used to send unsolicited email, nor will it be furnished to third parties. Please refer to The Journal of Bone & Joint Surgery Inc.'s Privacy Policy for further information.

Copyright © The Journal of Bone & Joint Surgery Inc. All rights reserved.
Share
Get Citation

Krishna Reddi Boddu Siva Rama, Sunil Apsingi, Sharmila Poovali, Anand Jetti; Timing of Tourniquet Release in Knee ArthroplastyMeta-Analysis of Randomized, Controlled Trials. The Journal of Bone & Joint Surgery. 2007 Apr;89(4):699-705.

Download citation file:

RIS (Zotero)

EndNote

BibTex

Medlars

ProCite

RefWorks

Reference Manager

Copyright © The Journal of Bone and Joint Surgery, Inc.
Rights & Permissions
Article Alerts

Processing your request... Please Wait.
Submit a Comment
Data Supplements