Abstract
Simultaneous bilateral TKA (SBTKA) in a single sitting is an attractive option for medically fit patients, with end-stage osteoarthritis (OA) of both the knees. It is a cost-effective procedure but is associated with increased blood loss and requirement for blood transfusion. We present a retrospective observational study of 144 patients who had SBTKA, with the mean age of 63.86 ± 7.38 years. We noted that all the cases of SBTKA would not require a blood transfusion, if the preoperative selection and optimization of the patient is done carefully and with the use of clean surgical technique and adequate thermocoagulation of the bleeders and perioperative use of tranexemic acid (TA) is done in these cases. 2/3rd of our patients did not require any blood transfusion, after SBTKA. We found that preexisting Hypertension and Hypothyroidism were associated with increased blood loss. The use of TA was a useful adjunctive measure to reduce perioperative blood loss. According to other studies which were reviewed there was no significant difference in blood loss with or without the use of a tourniquet. However, tourniquet was used in all patients in our study. The pre-operative level of Hemoglobin was an important factor to contribute to the requirement of blood transfusion after SBTKA. We recommend blood transfusion if the post-operative Hemoglobin level is less than 8.0 gm./dl.
Keywords: Blood loss, Blood transfusion, Total knee arthroplasty, Arthroplasty, Haemoglobin
1. Introduction
Knee Osteoarthritis (OA) is one of the most commoncauses of debilitating pain in the elderly population, which results in patients undergoing Total Knee Replacement (TKA) surgery.1 OA is an age-related degenerative process which usually affects both the knees, either simultaneously or sequentially. TKA allivates the patient’s symptoms and restores the individual to his regular routine. Many individuals suffering from advanced OA eventually may require to get their knees replaced over time. TKA surgery carries the potential risk of blood loss resulting in the requirement of allogeneic blood transfusion, which in turn may increase the risk of prosthetic joint infection. Moreover, the availability of allogeneic blood and donors is challenging, and other potential complications like Deep Vein Thrombosis (DVT), thromboembolism, etc. has made the Orthopaedic surgeons reluctant to perform bilateral knee Replacement in a single setting.2,3
This retrospective study was conducted to assess the blood loss and need for blood transfusion in the patients undergoing simultaneous bilateral TKA (SBTKR). This study was conducted on asian patient population and compared with the studies conducted in the other countries.
2. Material and methods
This was a retrospective observational study of 144 patients who underwent SBTKR, in a single sitting at a tertiary care center in India, from 2015 to 2018. All the patients with a clinico-radiological diagnosis of advanced (KL grade IV) OA of both the knees were included in this study. The patients with rheumatoid arthritis and other inflammatory arthritis, coagulopathy, and with a history of immunosuppressionwere excluded from this study. Anesthesia choice was left for the anesthetist based on patients fitness and preferences. In cases where spinal anesthesia was given, an epidural was added for postoperative pain management. The operative procedure was performed by a single Orthopaedic surgeon following a standard surgical protocol. The knee which was more painful was operated first. In cases where knee pain was similar, there left side was done first due to surgeon’s preference. The surgeon preferred to stand on the side of the limb getting operated. Single implant system were used in all the patients. Intravenous 1 gm tranexamic acid (TA) was given in 115 individuals, and a pneumatic tourniquet was used after limb elevation for 2 min, in all the patients. The pneumatic pressure was set at 150 mm Hg, above the systolic blood pressure. The tourniquet was inflated before the incision and was deflated after the cement was set in. No exsanguination of the limb was done, before inflation of the tourniquet. A single dose of TA was administered after completing the bone cuts in 89 patients (10 mg/kg), and three doses were administered in 26 patients (1st dose after taking all the bony cuts, second dose 3 h later and the third dose after 6 h of the first dose). Bilateral suction drains were inserted in all the patients, for 24–48 h. Preoperative and postoperative hemoglobin was tested in all the individuals, and a record of all comorbid illness was evaluated. Post-operative blood loss was monitored by the volume of the collection in the suction drains.
Patient’s data was divided into demographic data, preoperative hemoglobin, postoperative hemoglobin, comorbidities, drain output, tourniquet time, and total doses of TA. Post-operative hemoglobin value was obtained on day one following surgery. Patients were also evaluated for postoperative wound complications and signs of deep vein thrombosis.
Pharmacological thromboprophylaxis was given in the form of tablet Aspirin 75 mg (an antiplatelet drug), once daily for six weeks. This started on post operative day 1. Mechanical thromboprophylaxis in the form of calf pumps was used in all the patients post-operatively, after drain removal and dressing change above knee compression stockings were also used.
The tolerance to blood transfusion was kept to hemoglobin levels below 8 gm/dl. No other blood conservation methods such as cell-saver or autologous blood transfusion, etc. were used in our patients. Patients were followed up for one year. Institutional ethics committee approval was taken, and informed written consent was obtained from each.
3. Results
A total of 144 patients were included in this study, with the mean age of 63.86 ± 7.38 years (Table 1). Out of a total of 144 patients, 34 were males, and 110 were females. Fifty patients (34.7%) received an autologous blood transfusion in the postoperative period (11 Males and 39 Females). The mean age of these patients was 63.78 ± 7.67 years, which was comparable to the mean age of the total patients. Remaining 94 patients (65.3%) did not receive any post-operative blood transfusion (23 Males and 71 Females), the mean age of patients in this group was 63.95 ± 7.099 years which was also comparable to the total mean age of the study population. The two groups thus formed; Transfusion group (TG) and No transfusion group (NTG) were comparable in terms of age and sex distribution they only differed in the total number of patients in each group (Table 2).
Table 1.
Demographics and Statistics of this case series.
| Demographics and Statistics |
p-value | ||||
|---|---|---|---|---|---|
| Number of cases | Mean | Std. Dev. | |||
| AGE (in years) | Non-transfusion | 94 | 63.95 | 7.099 | .896 |
| Transfusion | 50 | 63.78 | 7.675 | ||
| PRE-OPERATIVE Haemoglobin (in gm/dl) | Non-transfusion | 94 | 13.06 | 1.181 | .000 |
| Transfusion | 50 | 11.71 | 1.172 | ||
| POST-OPERATIVE Haemoglobin (in gm/dl) | Non-transfusion | 94 | 9.87 | 1.089 | .000 |
| Transfusion | 50 | 7.67 | 0.67230 | ||
| DRAIN-RIGHT (in ml) | Non-transfusion | 94 | 423.33 | 220.372 | .497 |
| Transfusion | 50 | 396.02 | 244.925 | ||
| DRAIN-LEFT (in ml) | Non-transfusion | 94 | 419.15 | 206.087 | .872 |
| Transfusion | 50 | 425.62 | 266.223 | ||
| TOTAL-DRAIN (in ml) | Non-transfusion | 94 | 842.48 | 387.993 | .776 |
| Transfusion | 50 | 821.64 | 468.702 | ||
| TOURNIQUET-LEFT (in minutes) | Non-transfusion | 94 | 44.20 | 7.338 | .708 |
| Transfusion | 50 | 44.70 | 8.046 | ||
| TOURNIQUET-RIGHT (in minutes) | Non-transfusion | 94 | 44.97 | 8.453 | .276 |
| Transfusion | 50 | 43.30 | 9.179 | ||
| TOTAL-TOURNIQUET (in minutes) | Non-transfusion | 94 | 89.17 | 12.673 | .609 |
| Transfusion | 50 | 88.00 | 13.667 | ||
Table 2.
Correlation of blood transfusion to various parameters.
| BLOOD TRANSFUSION |
p-value | |||
|---|---|---|---|---|
| No-transfusion (NTG) | Transfusion (TG) | |||
| SEX | Male | 23 | 11 | 0.740 |
| Female | 71 | 39 | ||
| COMORBIDITIES | HTN | 46 | 14 | 0.003 |
| HTN + DM | 14 | 9 | ||
| Others | 12 | 18 | ||
| TRANEXAMIC ACID | No Dose | 16 | 13 | 0.422 |
| 1 Dose | 61 | 28 | ||
| 3 Dose | 17 | 9 | ||
| ASA | 2 | 91 | 46 | 0.201 |
| 3 | 3 | 4 | ||
The overall mean pre-operative hemoglobin value in the study population was 12.39 ± 1.18 gm/dl. The value of mean pre-operative hemoglobin level in the TG was 11.71 ± 1.17 gm/dl, and in NTG, it was 13.06 ± 1.18 gm/dl (Fig. 1). There was a difference between the values of the pre-operative hemoglobin levels in these two groups (Table 3), and this difference was statistically significant (P value < 0.005). The significance between the hypothyroidism and blood transfusion was found highly significant (p < 0.001).
Fig. 1.
Comparison of Haemoglobin level in Transfusion vs. Nontransfusion groups.
Table 3.
Comparison of Transfusion and Non Transfusion group, on various paprameters.
| NON TRANSFUSION GROUP (NTG) | TRANSFUSION GROUP (TG) |
CHI-SQUARE | df | P-VALUE | ||
|---|---|---|---|---|---|---|
| Tranexamic Acid | 1 dose | 69 | 36 | 0.33 | 1 | 0.857 |
| Tranexamic Acid | 3 doses | 25 | 14 | |||
| DM | Absent | 78 | 38 | 1.015 | 1 | 0.314 |
| Present | 16 | 12 | ||||
| Hypothyroidism | Absent | 83 | 26 | 24.051 | 1 | <0.001 |
| Present | 8 | 21 | ||||
| Hypertension | Absent | 86 | 38 | 6.548 | 1 | 0.011 |
| Present | 8 | 12 |
The cutoff value of 8.0 gm/dl for the postoperative hemoglobin level was taken as the tolerance for blood transfusion. The patients who had hemoglobin levels below 8 gm/dl received a blood transfusion in this study. The overall mean post-operative hemoglobin level in TG and NTG were 7.67 ± 0.678 gm/dl and 9.87 ± 1.08 gm/dl respectively.
A mean combined tourniquet time of the overall study population was 88.58 ± 13.17 min. There was no statistically significant difference in the mean combined tourniquet time in both the groups; in TG the mean was 88.00 ± 13.67 min and in NTG, it was 89.17 ± 12.67 min (P-value = 0.609).
The surgical drain was placed in all the patients, which was removed on the second postoperative day. The total amount of drain collection was recorded over two days. The mean value of the combined drain collections (from both right as well as a left limb) was 832.06 ± 428.34 ml in the whole study population. The mean value of drain collection in the TG was 821.64 ± 468.70 ml (Right side- 396.02 ± 244.9 ml, Left side- 425.62 ± 266.22 ml) and in the NTG it was 842.48 ± 387.99 ml (Right side- 423.33 ± 220.34 ml, left side- 419.19 ± 206.08 ml), there was no statistically significant difference in the drain collection between the two groups (P value - 0.776).
We recorded some comorbid illnesses in our patient population like hypertension (HTN) in 83 individuals (HTN alone in 60, HTN with Diabetes Mellitus (DM) in 23) and Hypothyroidism in 19 patients. There was a statistically significant correlation between HTN and the need for post-operative blood transfusion. No patients were had postoperative wound complications and signs of deep vein thrombosis. Out of a total of 50 patients in the TG, 23 patients (46%) had HTN. A total of 29 patients in our study suffered from Hypothyroidism out which 21 patients received a post-operative blood transfusion, and eight patients did not receive a transfusion, which was statistically significant (p < 0.001).
In this study, the patients received Tranexamic acid (TA) TA heterogeneously and therefore, acted as comparative groups. Some received a single dose (61.7%) or three doses (18.05%) of TA, and some patients (20.25%) did not receive any TA. Out of 89 patients who received a single dose of TA, 28 patients (31.4%) required a blood transfusion in the postoperative period, and out of 26 patients who received three doses of TA, nine patients (34.6%) required transfusion. We did not find any statistically significant difference in need of transfusion between the patients who received single or three doses of TA. Out of 29 patients who did not receive any TA, 13 (44.8%) patients required transfusion; there was a noticeable difference in the percentage of patients who required transfusion after receiving TA and those who did not receive it. However, this difference was not found to be statistically significant (P value = 0.422).
4. Discussion
Many surgeons are hesitant in doing SBTKR, in a single setting because of the fear of increased blood loss requiring blood transfusion and other complications.4,5 Bould et al.6 suggested that hypothermia, prolonged tourniquet application, and trauma are the main causes of increased bleeding in another limb while operating for SBTKR in a single setting. Hence, we tried decreasing the tourniquet timing and hypothermia in our cases. Knee OA is usually bilateral disease, and the patient who undergoes unilateral TKA may subsequently need a contralateral knee Replacement. SBTKR, in a single sitting, is, therefore, an attractive option for these patients with bilateral affection of the knees. It is reported to be a better option in terms of knee score and survivorship as compared to sequential TKA.7 Low levels of preoperative hemoglobin in patients undergoing SBTKR in a single setting has shown 17 times increase in chances of postoperative anemiaand need of blood transfusion.8
The use is of TA is effective in controlling blood loss, which usually peaks after the release of a pneumatic tourniquet.9 A single dose of TA has been shown effective in reducing blood loss during TKA surgery. We did not find any statistically significant difference between a single dose and three doses of TA administration. TA is a key factor which can encourage orthopedic surgeon to opt for SBTKR in a single setting, as it reduces overall blood transfusion requirements.10,11 In our study, there was a difference but not statistically significant due to smaller sample size. The single dose or three doses did not result in any major change and was statistically insignificant. Hencea single preoperative dose of TA was found adequate.
The overall transfusion rate in our study was 34.7%, and about 2/3rd cases did not require a blood transfusion after SBTKR. We found that pre-operative hemoglobin and comorbidities like HTN had a significant relationship with post-operative transfusion. There are studies which suggest that tourniquet may cause many associated complications like post-operative pain, excessive bleeding after the release of tourniquet, and wound complications.12,13 On the contrary, many authors suggest that there are certain advantages of tourniquet such as reduced intraoperative bleeding and clear visualization during surgery in turn reducing operative time. In our study, the tourniquet was applied to all the limbs which were operated, and we did not observe any significant tourniquet associated complication. So a limited time application of a pneumatic tourniquet seems safe in TKA surgery.
Dy CJ et al.14 in their meta-analysis demonstrated the importance of identifying the cardiovascular co-morbidities before planning bilateral TKA, so to prevent severe cardiovascular events, postoperatively. We did a dobutamine stress echo (DSE) for all patients planned for SBTKR. This fact was taken into account in all of our patients. Adili et al.15 demonstrated that SBTKR in a single setting in a patient aged more than 75 years showed good functional outcome and satisfaction as compared to sequential Replacement of both the knees despite the fact of increase chances of cardiovascular complication in these age groups. The senior author (RV) had earlier reported a series of 46 patients of SBTKR, in the patients above 70 years of age, where no perioperative mortality was encountered.16 Parvizi et al.17 found 30-day mortality rates from 0.17% to 0.49% undergoing SBTKR, but these studies were carried out more than a decade ago, and the current medical science has progressed since then, decreasing these rates. In the present study, we did not encounter either any major cardiovascular complications or mortality in the perioperative period.
Some studies imply the use of surgical drain to be the etiology of infection and blood transfusion in patients who are undergoing Replacement.18 However, we did not encounter any drain related complication despite applying a tourniquet to all the operated limbs and using surgical drains. Stubbs et al. in his study concluded that SBTKR is a good option for a patient without associated postoperative complications when compared with a single sitting.19 Our study corroborates with findings of Stubbs et al. in terms of post-operative complication although, we did not do a clinical evaluation for the functional outcome. Bullock et al.20 observed that perioperative complications, including myocardial infarction, postoperative confusion were higher in SBTKR and concluded that the mortality rates, compared to unilateral and bilateral TKA were the same in thirty day and one-year data analysis. Patients who had associated cardiovascular comorbidities like HTN have shown to have increased risk of blood transfusion in the previous studies.21,22
We did not find any literature on the relation between hypothyroidism and increase the need for allogenic blood transfusion after Orthopaedic surgery. However, in our study the co-relation was markedly significant. Hence, we are unable to derive any conclusion on this fact and suggest further studies to justify whether there is any relationship between hypothyroidism and increased need for postoperative blood transfusion.
Postoperative infection is a nightmare for an Orthopaedic surgeon, which makes them reluctant to opt for SBTKR. Hillge et al.23 in their meta-analysis demonstrated an increased risk of bacterial infection in allogeneic blood transfusion in trauma patients due to immune hypo-responsiveness, but the same is not valid for the elective Orthopaedic procedures, like Replacement. We also did not find any increased incidence of bacterial infections in our case series of patients who had received an allogeneic blood transfusion after SBTKR.
We realize some shortcomings and limitations of this study, as follow:
-
1.
This being a retrospective study is not as good as a prospective study.
-
2.
The number of patients is comparatively lesser in this series, but we believe these are enough to reach a logical conclusion.
-
3.
An only a limited number of factors influencing blood loss and the need for blood transfusion were included and analyzed. Some other reported factors like Obesity etc. might have given more insight.
-
4.
Correlation of blood loss and function outcomes might have also given useful information.
-
5.
The study of the association of chronic use of antiplatelet drugs and total blood loss could have provided been useful information.
5. Conclusion
Simultaneous bilateral TKA (SBTKR) in a single sitting is an attractive option for medically fit patients, with end-stage OA of both the knees. It is a cost-effective procedure but is associated with increased blood loss and requirement for blood transfusion. However, all the cases of SBTKR would not require a blood transfusion, if the preoperative selection and optimization of the patient is done carefully and with the use of clean surgical technique and adequate thermocoagulation of the bleeders and perioperative use of TA is done in these cases. About 2/3rd of our patients did not require any blood transfusion, after SBTKR. We found that preexisting Hypertension and Hypothyroidism were associated with increased blood loss. The use of TA was a useful adjunctive measure to reduce perioperative blood loss. There was no significantdifference in blood loss with or without the use of a tourniquet. The pre-operative level of Hemoglobin was an important factor to contribute to the requirement of blood transfusion after SBTKR. We recommend blood transfusion if the post-operative Hemoglobin level is less than 8.0 gm./dl.
Funding
There is no funding source.
Ethical approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Declaration of competing interest
The authors declare that they have no conflict of interest.
References
- 1.Nichols C.I., Vose J.G. Comparative risk of transfusion and incremental total hospitalization cost for primary unilateral, bilateral, and revision total knee replacement procedures. J. Replacement. 2016;31(3):583–589. doi: 10.1016/j.arth.2015.10.013. [DOI] [PubMed] [Google Scholar]
- 2.Odum S.M., Troyer J.L., Kelly M.P., Dedini R.,D., Bozic K.J. A cost-utility analysis comparing the cost-effectiveness of simultaneous and staged bilateral total knee replacement. J Bone Joint Surg. 2013;95(16):pp1441–p1449. doi: 10.2106/JBJS.L.00373. [DOI] [PubMed] [Google Scholar]
- 3.Jaffer A.K., Barsoum W.K., Krebs V., Hurbanek J.G., Morra N., Brotman D.J. Mayo Clinic Proceedings. Elsevier; 2005. Duration of anesthesia and venous thromboembolism after hip and knee replacement; pp. 732–738. [DOI] [PubMed] [Google Scholar]
- 4.Hendrickson J.E., Roubinian N.H., Chowdhury D. Incidence of transfusion reactions: a multicenter study utilizing systematic active surveillance and expert adjudication. Transfusion. 2016;56(10):pp2587–p2596. doi: 10.1111/trf.13730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Zhu Y., Zhang F., Chen W., Liu S., Zhang Q., Zhang Y. Risk factors for periprosthetic joint infection after total joint replacement: a systematic review and meta-analysis. J Hosp Infect. 2015;89(2):82–89. doi: 10.1016/j.jhin.2014.10.008. [DOI] [PubMed] [Google Scholar]
- 6.Bould M., Freeman B.J., Pullyblank A., Newman J.H. Blood loss in sequential bilateral total knee replacement. J Replacement. 1998;13(1):pp77–79. doi: 10.1016/s0883-5403(98)90078-x. [DOI] [PubMed] [Google Scholar]
- 7.Walmsley P., Murray A., Brenkel I.J. The practice of bilateral, simultaneous total knee Replacement in Scotland over the last decade. Data from the Scottish Replacement Project. Knee. 2006;13(2):pp102–105. doi: 10.1016/j.knee.2006.01.003. [DOI] [PubMed] [Google Scholar]
- 8.Lane G.J., Hozack W.J., Shah S. Simultaneous bilateral versus unilateral total knee replacement. Outcomes analysis. Clin Orthop Relat Res. 1997;345:pp106–112. [PubMed] [Google Scholar]
- 9.Good L., Peterson E., Lisander B. Tranexamic acid decreases external blood loss but not hidden blood loss in total knee Replacement. Br J Anaesth. 2003;90(5):596–599. doi: 10.1093/bja/aeg111. [DOI] [PubMed] [Google Scholar]
- 10.Gandhi R., Evans H.M.K., Mahomed S.R., Mahomed N.N. Tranexamic acid and the reduction of blood loss in total knee and hip replacement: a meta-analysis. BMC Res Notes. 2013;7(6):184. doi: 10.1186/1756-0500-6-184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kakar P.N., Gupta N., Govil P., Shah V. Efficacy and safety of tranexamic acid in control of bleeding following TKA: a randomized clinical trial. Indian J Anaesth. 2009;53(6):667–671. [PMC free article] [PubMed] [Google Scholar]
- 12.Saunders K.C., Louis D.L., Weingarden S.I., Waylonis G.W. Effect of tourniquet time on postoperative quadriceps function. Clin Orthop Relat Res. 1979;143:pp194–p199. [PubMed] [Google Scholar]
- 13.Lee Y.H., Wei F.C., Lee J., Su M.S., Chang Y.C. Effect of postischemic reperfusion on microcirculation and lipid metabolism of skeletal muscle. Microsurgery. 1995;16(8):522–527. doi: 10.1002/micr.1920160803. [DOI] [PubMed] [Google Scholar]
- 14.Dy C.J., Wilkinson J.D., Tamariz L., Scully S.P. Influence of preoperative cardiovascular risk factor clusters on complications of total joint replacement. Am J Orthop (Belle Mead NJ) 2011;40(11):560–565. [PubMed] [Google Scholar]
- 15.Adili A., Bhandari M., Petruccelli D., De Beer J. Sequential bilateral total knee replacement under 1 anesthetic in patients > or = 75 years old: complications and functional outcomes. J Replacement. 2001;16(3):271–278. doi: 10.1054/arth.2001.21495. [DOI] [PubMed] [Google Scholar]
- 16.Vaishya R., Vijay V., Mani K.K.C., Agarwal A.K. Is simultaneous bilateral total knee replacement safe in geriatric population? A retrospective cohort study with upto 9 years follow up. J Clin Orthop Trauma. 2018;9(2):107–111. doi: 10.1016/j.jcot.2017.05.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Parvizi J., Sullivan T.A., Trousdale R.T., Lewallen D.G. Thirty-day mortality after total knee replacement. J Bone Joint Surg. 2001;83(8):1157–1161. doi: 10.2106/00004623-200108000-00004. [DOI] [PubMed] [Google Scholar]
- 18.Demirkale I., Tecimel O., Sesen H., Kilicarslan K., Altay M., Dogan M. Nondrainage decreases blood transfusion need and infection rate in bilateral total knee replacement. J Replacement. 2014;29(5):pp993–p997. doi: 10.1016/j.arth.2013.10.022. [DOI] [PubMed] [Google Scholar]
- 19.Stubbs G., Pryke S.E.R., Tewari S. Safety and cost benefits of bilateral total knee Replacement in an acute hospital. ANZ J Surg. 2005;75(9):739–746. doi: 10.1111/j.1445-2197.2005.03516.x. [DOI] [PubMed] [Google Scholar]
- 20.Bullock D.P., Sporer S.M., Shirreffs T.G., Jr. Comparison of simultaneous bilateral with unilateral total knee replacement in terms of perioperative complications. J Bone Joint Surg. 2003;85(10):1981–1986. doi: 10.2106/00004623-200310000-00018. [DOI] [PubMed] [Google Scholar]
- 21.Nwachukwu B.U., Collins J.E., Nelson E.P., Concepcion M., Thornhill T.S., Katz J.N. Obesity & hypertension are determinants of poor hemodynamic control during total joint replacement: a retrospective review. BMC Muscoskel Disord. 2013;14 doi: 10.1186/1471-2474-14-20. pp 14:20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Leonard L., Williamson D.M., Ivory J.P., Jennison C. An evaluation of the safety and efficacy of simultaneous bilateral total knee replacement. J Replacement. 2003;18(8):972–978. doi: 10.1016/s0883-5403(03)00282-1. [DOI] [PubMed] [Google Scholar]
- 23.Hill G.E., Frawley W.H., Griffith K.E., Forestner J.E., Minei J.P. Allogeneic blood transfusion increases the risk of postoperative bacterial infection: a meta-analysis. J Trauma. 2003;54(5):908–914. doi: 10.1097/01.TA.0000022460.21283.53. [DOI] [PubMed] [Google Scholar]