a Departamento de Ortopedia, Santa Casa de São Paulo, São Paulo, SP, Brasil
b Serviço de Ortopedia e Traumatologia, Hospital Israelita Albert Einstein, Universidade Federal de São Paulo (Unifesp), São Paulo, SP,Brasil
c Instituto de Ortopedia, Hospital das Clínicas, Faculdade de Medicina, Universidade de São Paulo (USP), São Paulo, SP, Brasil
d Escola Paulista de Medicina, Universidade Federal de São Paulo (Unifesp), São Paulo, SP, Brasil
Introduction
Total knee arthroplasty (TKA) is one of the surgical proceduresmost indicated for treating osteoarthrosis,1which is a diseaseof inflammatory and degenerative nature that causes destruc-tion of the knee joint cartilage and leads to joint deformity.2TKA is an increasingly common procedure,3-6but it is asso-ciated with major postoperative blood losses that may reachquantities greater than or equal to 1.5 L. In such cases, trans-fusion becomes inevitable.7,8The bleeding occurs mainly afterremoval of the tourniquet.9,10
It is understood that in cases of trauma and in large-scale surgical procedures like TKA, in which there is acuteblood loss, oxygenation is the main indicator regarding bloodvolume replacement. This is a decisive moment, at whichreplacement should be performed using blood or blood com-ponents, rather than by means of acellular solutions alone.Thus, transfusions can be done using homologous blooddonated by another person, or using autologous blood, inwhich the donor and recipient are the same person. In mostcases, transfusion using homologous blood is done moreoften.11
However, if the indications for transfusions are analyzed,it is seen that there is no consensus regarding what the mini-mum acceptable value for the hemoglobin level would be, foradequate tissue perfusion to be maintained. Discrepant val-ues have been reported, ranging from levels as low as 1.8 g/dLto normal levels like 12 g/dL.
The present study had the aim of analyzing the relation-ship between the hematimetric variation and the presence of clinical symptoms of hypoperfusion, for indicating bloodtransfusion in patients undergoing TKA.
Materials and methods
A retrospective analysis was conducted on data gathered fromthe medical files of 55 patients who underwent unilateral TKA,performed at the orthopedics and traumatology service of ahospital between February 2011 and December 2012. Amongthese patients, 35 (63.63%) were female and 20 (36.36%) weremale, and the mean age was 68.3 years (range: 45-86), as canbe seen in Fig. 1.
All the procedures were performed by a team of surgeonswith experience of the surgical technique of unilateral TKA.
As inclusion criteria, the patients needed to present pri-mary gonarthrosis with an indication for TKA, undergo apreoperative cardiological assessment, remain within cate-gories I-III of the American Society of Anesthesiology (ASA),present absence of blood dyscrasia and have preoperative
hemoglobin measurements available. No minimum hemati-metric value was established for the surgery: only clinicalcriteria for blood perfusion. Patients who were regularly usingmedications with the potential to alter their coagulation wereadvised to suspend their use ten days before the surgicalintervention, and all of the patients presented normal coag-ulograms on the day of the procedure. Any intraoperative orpostoperative complication was considered to be an exclusioncriterion.
The antibiotic cefazolin was used as a single dose of 1 gintravenously, to induce anesthesia. Prevention of venousthromboembolism followed the guidance of the venousthromboembolism prevention committee, with subcutaneousadministration of 40 mg of enoxaparin per day, starting 6 hafter onset of the anesthetic block and continuing for ten days.In all the patients, the anesthesia consisted of a locoregionalblock and the criterion for transoperative and immediatepostoperative volume replacement was restrictive, compris-ing crystalloid infusion at the rate of 10 mL/kg/h. Asepsis andantisepsis were performed, with application of specific topicalsolutions, and placement of appropriate surgical fields.
During the surgical procedure, a pneumatic tourniquetwith a pressure of 375 mmHg was used. This was positionedat the root of the thigh, which was protected with orthopediccotton wool and a crepe bandage. The conventional techniquefor total knee arthroplasty was followed, by means of a medianaccess route and medial parapatellar arthrotomy, eversion ofthe patella, disturbance of the knee extensor mechanism, jointdislocation and extensive violation of the soft tissues sur-rounding the joint.13-15All the patients received a cementedprosthesis, which is used more frequently for technical andeconomic reasons.16
After the pneumatic tourniquet had been released,hemostasis was performed and a one-way vacuum suctiondrain was placed inside the joint, with its outlet in the lateraldistal region of the femur. Suturing was performed in layers,with complete flexion-extension movements made after clos-ing each layer. Dressings were applied from the inguinal regionto the foot, consisting of successive layers of compressive ban-dages and orthopedic cotton wool, followed by a plaster-castsplint extending from the inguinal region to the malleolus.
Volume replacement with crystalloids was performed inthe ward, with observance of a ration of 3:1 for the blood lossescollected through the vacuum drain and the estimated dailyloss into the dressings. Hemoglobin control was applied 24 hafter the end of the surgical procedure, given that this is theideal time, since it corresponds to the time of greater volumedrained through the drain collector system. After this time,the volume becomes insignificant.17
The length of hospital stay was three days. Isometric andisotonic exercises were performed on the entire lower limb,beginning on the first postoperative day and continuing untildischarge from hospital. The aims were to achieve active flex-ion of at least 90?, complete extension and walking with partialweight-bearing, with the aid of a walking frame.
Results
In evaluating the hemoglobin levels after 24 h, it was observedthat the mean loss was 3.33 g/dL (24.78%), with a range from 0.5
to 6.3 g/dL. The postoperative hemoglobin assay ranged from7.5 to 12.7 g/dL, and most of the patients (67.27%) presentedlevels between 9 and 12 g/dL (Table 1).
To view the data presented in Table 1 better, a histogram offrequencies of numbers of patients was constructed accordingto the postoperative hemoglobin assay values (Fig. 2).
The percentage reduction in hemoglobin level ranged from5.55% to 40.91%. Most of the patients (36) presented a decline,which ranged from 15% to 30%. All the patients who receivedtransfusions presented percentages greater than 20%. How-ever, 39 patients presented percentage losses greater than20% and only six (15.38%) received transfusions, as shown inTable 2.
The total number of patients and number of transfusedpatients according to the percentage decline in hemoglobinlevels after the operation can be seen in Fig. 3.
Six patients (10.9%) underwent exclusively homologousblood transfusion, because they presented symptoms of tis-sue hypoperfusion, such as persistent hypotension, loss ofconsciousness, cold sweating, coercible vomiting and men-tal confusion. The criteria of hematimetric variation werenot determinants for the transfusions. However, in analyz-ing the values, we observed that all the transfused patients
presented postoperative hemoglobin counts that ranged from7.5 to 8.8 g/dL (Table 3). It should be emphasized that otherpatients with postoperative hemoglobin counts lower than8.8 g/dL did not have the need for transfusion, since they didnot present clinical symptoms of hypoperfusion (Table 1).
Discharge from hospital was delayed by one day for thepatients who underwent transfusions. However, no othercomplications were observed.
The behavior of the patients who underwent blood trans-fusion, regarding their pre and postoperative hemoglobinlevels and the percentage reductions, can be seen better inFig. 4.
Discussion
Blood loss during TKA may lead patients to the need toundergo transfusion. However, there is no consensus in thepertinent literature regarding the indications for transfusionaccording to hemoglobin levels.
It is worth recalling the observation made by Vuille-Lessardet al.18that doctors do not have definitive indications orsupport guidelines for the need for blood transfusion afterorthopedic surgery, which ultimately results in a diversity oflevels used in transfusion practice.
In analyzing blood management in patients undergoingtotal knee or hip arthroplasty, Bierbaum et al.19stated thattheir study was the first to prospectively assess the role of hemoglobin according to its levels or specific categories. Theyevaluated the need for autologous or homologous transfusionbased on the initial hemoglobin level. Patients whose pre-operative hemoglobin level was 13 g/dL or less would needa homologous blood transfusion, particularly those whoseinitial hemoglobin level was between 10 and 13 g/dL. Sub-sequently, Billote et al.20made a complete review of themanuscript of Bierbaum et al.19and disagreed with the origi-nal result, with the justification that the category of patientswith initial hemoglobin levels from 10 to 13 g/dL presented inthat study was very broad, in that it mixed anemic and non-anemic elderly individuals and other eligible donors. Also inBillote et al.,20Bierbaum et al. replied and stated that fromthe conceptual point of view, the interval of 10-13 g/dL in theirstudy was based on information from the World Health Orga-nization (WHO). Bierbaum et al.19agreed with Billote et al.20in stating that additional investigations of blood managementstrategies in these surgical procedures were necessary.
In this regard, Ng et al.21made reference to the studiesof Salido et al.22and Hatzidakis et al.,23and stated only thatpatients with preoperative hemoglobin levels less than 13 g/dLare four to six times more likely to need transfusion than arepatients with hemoglobin levels between 13 and 15 g/dL and15 times more likely than those with a level of 15 g/dL.
In the present study, it was observed that among thepatients who underwent blood transfusion, only one ofthem presented a preoperative hemoglobin level lower than11 g/dL, while the others (total of five) presented preoperative
hemoglobin levels between 12.6 and 13.1 g/dL. This was notthe decisive factor for indicating transfusion.
Regarding postoperative hemoglobin levels, in anotherstudy24on the use of tourniquets in TKA cases in two groupsof patients (with and without a tourniquet), it was reportedthat two patients in each group underwent blood transfu-sion after the operation. One patient underwent transfusionwhen a hemoglobin level of 8.3 g/dL was presented, accompa-nied by tachycardia and mild dyspnea, while the other threepresented low hemoglobin levels (7.2, 7.6 and 7.7 g/dL). Themean decrease in hemoglobin level for these four patients was4.1 g/dL.
In the present study, it was observed that all the trans-fused patients presented postoperative hemoglobin counts ofbetween 7.5 and 8.8 g/dL. However, other patients with post-operative hemoglobin counts lower than 8.8 g/dL did not needtransfusion, given that they did not present clinical symptomsof hypoperfusion and the hematimetric variation criteria werenot determinant for the transfusions.
Other authors25who studied transfusions of autologousand homologous blood in TKA surgery cases made com-parisons through distributing the patients into two groups:those who received homologous blood transfusions and thosewho received autologous blood transfusions. They empha-sized that the indication for transfusion for both groupswas when the patient presented a postoperative hemoglobinlevel of 8.5 g/dL. They also pointed out that their study hadsome limitations and explained that although the recom-mendation from local studies was 8.5 g/dL, this was notalways followed by some surgeons, who were still reluc-tant to apply this because they thought that it was toolow.
These authors25took the view that this was the expla-nation for higher hemoglobin levels in the homologoustransfusion group, which was not seen in the autologous groupbecause the protocol for this was controlled by the techniqueand the anesthesia by the anesthesiologist.
In the present study, the view taken was that, unlike inthe cases of the above authors,25the patients who underwenthomologous blood transfusion received this solely because becausethey presented symptoms of tissue hypoperfusion. The hema-timetric variation criteria were not determinants for thetransfusions.
In a comparative study on blood transfusion practice inelective surgical procedures in three hospitals in France,Vuille-Lessard et al.18concluded that in general, the patientsreceived transfusions when they presented hemoglobin levelsof between 7.5 and 8.0 g/dL. There were significant differencesbetween the hospitals, regarding the distribution of the proce-dures, hemoglobin levels and blood losses, and also in relationto the decrease in hemoglobin levels and duration of this state.These authors stated that in 85% of the transfusions, only thehemoglobin level was envisaged.
In the present study, the majority of the patients (67.27%)presented postoperative hemoglobin levels between 9 and12 g/dL, and none of them required transfusion. However,among the 11 patients with postoperative hemoglobin levelslower than 9 g/dL, six (54.54%) required transfusion. In otherwords, patients with hemoglobin levels lower than 9 g/dL fol-lowing TKA surgery presented twice as much possibility ofneeding blood transfusion. This hemoglobin value was shownto be a minimum threshold for possible transfusion, in relationto clinical symptoms.
Also in the present study, the majority of the patients(65.45%) presented percentage decreases in hemoglobin lev-els ranging from 15% to 30%. However, among the transfusedpatients alone, the mean percentage was greater (33.84%),with a range from 24.77% to 40.47%.
In total, there were 39 patients with a change of more than20%, but only 15.38% required transfusion. Thus, it seemsthat indicating blood replacement according to the percent-age hematimetric loss is not an appropriate criterion for usein clinical practice, since no standard that would or would notjustify blood transfusion was found.
Conclusion
When patients present declines in hemoglobin count greaterthan 20% and values lower than 9 g/dL after surgery, this suggests a possible need for blood transfusion, which shouldonly be indicated when accompanied by greater symptoms oftissue hypoperfusion.
Conflicts of interest
The authors declare no conflicts of interest.
1. Cintra FF, Yepéz AK, Rasga MGS, Abagge M, Alencar PGC.Componente tibial na revisão da artroplastia do joelho:comparac¸ão entre fixac¸ão cimentada e híbrida. Rev BrasOrtop. 2011;46(5):585-90.2. Camanho GL. Tratamento da osteoartrose do joelho. Rev BrasOrtop. 2001;36(5):135-40.3. Iamaguchi MM, Helito CP, Gobbi RG, Demange MK, Tirico LEP,Pecora JR, et al. Valor da avaliac¸ão radiográfica pré-operatóriados defeitos ósseos no joelho nas revisões de artroplastia. RevBras Ortop. 2012;47(6):714-8.4. Khatod M, Inacio M, Paxton EW, Bini SA, Namba RS, BurchetteRJ, et al. Knee replacement: epidemiology, outcomes, andtrends in Southern California: 17,080 replacements from 1995through 2004. Acta Orthop. 2008;79(6):812-9.5. Wells VM, Hearn TC, McCaul KA, Anderton SM, Wigg AE,Graves SE. Changing incidence of primary total hiparthroplasty and total knee arthroplasty for primaryosteoarthritis. J Arthroplasty. 2002;17(3):267-73.6. Pugely AJ, Martin CT, Gao Y, Mendoza-Lattes S, Callaghan JJ.Differences in short-term complications between spinal andgeneral anesthesia for primary total knee arthroplasty. J BoneJoint Surg Am. 2013;95(3):193-9.7. Lotke PA, Faralli VJ, Orenstein EM, Ecker ML. Blood loss aftertotal knee replacement. Effects of tourniquet release andcontinuous passive motion. J Bone Joint Surg Am.1991;73(7):1037-40.8. Singh VK, Singh PK, Javed S, Kumar K, Tomar J. Autologoustransfusion of drain contents in elective primary kneearthroplasty: its value and relevance. Blood Transfus.2011;9(3):281-5.9. Andrade MAP, Campos TVO, Silva BFA, Assis ME, Boechat LC,Biondi LF, et al. Avaliac¸ão prospectiva dos pacientessubmetidos a artroplastia total do joelho com e sem colocac¸ãode dreno de succ¸ão. Rev Bras Ortop. 2010;45(6):549-53.10. Burke DW, O'Flynn H. Primary total knee arthroplasty. In:Chapman MW, editor. Chapman's orthopedic surgery.Philadelphia: Lippincott Williams & Wilkins; 2001. p. 2870-95.11. Vane LA, Ganem EM. Doac¸ão homóloga versus autóloga esubstitutos da hemoglobina. In: Cavalcanti IL, Cantinho FAF, Assad AR, editors. Medicina perioperatória. Rio de Janeiro:Sociedade de Anestesiologia do Estado do Rio de Janeiro;2006. p. 291-306.12. Hajjar LA, Auler JOC Jr, Santos L, Galas F. Blood transfusion incritically ill patients: state of the art. Clinics.2007;62(4):507-24.13. Motta Filho GR, Cavanellas N. Artroplastia minimamenteinvasiva do joelho. Rev Bras Ortop. 2007;42(9):269-77.14. Scott RD, Volatile TB. Twelve years' experience with posteriorcruciate-retaining total knee arthroplasty. Clin Orthop RelatRes. 1986;(205):100-7.15. Malkani AL, Rand JA, Bryan RS, Wallrichs SL. Total kneearthroplasty with the kinematic condylar prosthesis: aten-year follow-up study. J Bone Joint Surg Am.1995;77(3):423-31.16. Cobra HAAB, Palma IM. Artroplastia total do joelho.SEMCAD/PROATO/SBOT - Sociedade Brasileira de Ortopedia eTraumatologia. Porto Alegre: Artmed/Panamericana; 2008.17. Lopes AIS, Cobra HAAB, Cavanellas NT. Avaliac¸ão da perdasanguínea em artroplastia total primária do joelho comdrenagem de succ¸ão: estudo prospectivo. Orto & Trauma:Discussões e Complicac¸ões. 2009;(6):4-6.18. Vuille-Lessard E, Boudreault D, Girard F, Ruel M, Chagnon M,Hardy JF. Red blood cell transfusion practice in electiveorthopedic surgery: a multicenter cohort study. Transfusion(Paris). 2010;50(10):2117-24.19. Bierbaum BE, Callaghan JJ, Galante JO, Rubash HE, Tooms RE,Welch RB. An analysis of blood management in patientshaving a total hip or knee arthroplasty. J Bone Joint Surg Am.1999;81(1):2-10.20. Billote DB, Bierbaum BE, Callaghan JJ, Galante JO, Rubash HE,Tooms RE, et al. Analysis of blood management. J Bone JointSurg Am. 2000;82(6):900-1.21. Ng VY, Lustenberger D, Hoang K, Urchek R, Beal M, CalhounJH, et al. Preoperative risk stratification and risk reduction fortotal joint reconstruction: AAOS exhibit selection. J Bone JointSurg Am. 2013;95(4):e191-215.22. Salido JA, Marín LA, Gómez LA, Zorrilla P, Martínez C.Preoperative hemoglobin levels and the need for transfusionafter prosthetic hip and knee surgery: analysis of predictivefactors. J Bone Joint Surg Am. 2002;84(2):216-20.23. Hatzidakis AM, Mendlick RM, McKillip T, Reddy RL, Garvin KL.Preoperative autologous donation for total joint arthroplasty.An analysis of risk factors for allogenic transfusion. J BoneJoint Surg Am. 2000;82(1):89-100.24. Tai TW, Chang CW, Lai KA, Lin CJ, Yang CY. Effects oftourniquet use on blood loss and soft-tissue damage in totalknee arthroplasty: a randomized controlled trial. J Bone JointSurg Am. 2012;94(24):2209-15.25. Sarkanovic ML, Gvozdenovic L, Savic D, Ilic MP, Jovanovic G.Autologous blood transfusion in total knee replacementsurgery. Vojnosanit Pregl. 2013;70(3):274-8.
