Total knee arthroplasty (TKA) is a successful intervention for the treatment of end-stage osteoarthritis.1,2 In 2014, a total of 643,732 TKA procedures were performed in the United States,3 and that number is expected to increase to 3.48 million by 2030.4 This rapid increase, however, carries significant cost considerations for the health care system. In 2013, joint replacement was the most common hospital procedure paid for by Medicare, accounting for nearly 450,000 inpatient admissions and $6.6 billion in payments.5 In the setting of increasing health care expenditures and limited resources, strategies to reduce costs without compromising patient safety are of paramount importance.
One strategy to minimize excessive costs is to question the utility of routine practices that may be of nominal value to patient outcomes and safety. For example, historically, the practice of obtaining postoperative complete blood count (CBC) testing after TKA has been routine for most arthroplasty surgeons. Knowing the utility of routine postoperative hemoglobin and hematocrit (H&H) values, which is the focus of the CBC, as well as the risks associated with eliminating these laboratory tests is particularly germane because of the surge of outpatient TKA procedures, in which laboratory tests are not performed on postoperative days.6,7 Moreover, the ubiquitous use of tranexamic acid (TXA) and the associated decrease in the prevalence of blood transfusions after TKA further brings into question the need for routine postoperative H&H in contemporary practice.8–10
Greco et al11 published a report concluding that routine postoperative CBC testing is unnecessary after total joint arthroplasty. In their study of 1132 total hip and knee arthroplasty procedures, 12 (1.06%) patients were transfused, all of whom had preoperative anemia. Only 0.27% (2 of 731) of their patients undergoing primary TKA required transfusion. These authors did not perform a financial impact analysis. Shaner et al12 reported that routine postoperative laboratory tests were unnecessary after unicompartmental knee arthroplasty. In their study of 322 consecutive patients who underwent unicompartmental knee arthroplasty, no patients required transfusion, and only 1.6% of patients required interventions for either hypokalemia or hyperglycemia. They concluded that routine postoperative laboratory tests after unicompartmental knee arthroplasty are unnecessary, especially considering that charges associated with these laboratory studies were more than $85,000 for their hospital. Kildow et al13 published similar findings for total hip arthroplasty (THA), concluding that CBC testing is unnecessary after THA for patients who are not anemic preoperatively and receive TXA. To the authors' knowledge, this study is the first to report on the financial effect of eliminating routine postoperative CBC testing for primary TKA.
The primary goal of this study was to determine the utility of obtaining routine postoperative CBC testing for patients undergoing primary TKA with concomitant use of intraoperative TXA. Specifically, the authors sought to determine (1) the percentage of patients undergoing TKA with TXA who had abnormally low postoperative H&H and (2) the percentage of patients who required an intervention based on these laboratory findings. The secondary goal was to determine (1) the total costs and the cost per intervention of postoperative CBC testing and (2) the potential cost savings associated with eliminating these laboratory tests.
Materials and Methods
The authors conducted a retrospective chart review of the electronic medical record at their institution. Institutional review board approval was obtained before initiation of the study. All patients undergoing TKA performed by a single surgeon (J.R.H.F.) at a single institution between May 1, 2016, and May 31, 2018, were identified from a prospectively collected database. Inclusion criteria included all patients undergoing primary unilateral TKA who received TXA. Exclusion criteria included revision arthroplasty, unicompartmental arthroplasty, patellofemoral arthroplasty, bilateral simultaneous TKA, and the absence of preoperative or postoperative H&H values.
All patients underwent mini-midvastus arthrotomy. Standard Bovie (ValleyLab; Medtronic) electrocautery was routinely used, and no bipolar sealers (ie, Aquamantys; Medtronic) were used. Tourniquets were used and deflated before closure of the arthrotomy. Drains were not used. All patients received TXA. Between May 1, 2016, and February 28, 2018, all patients who underwent this procedure received 1 g of intravenous (IV) TXA before incision and 1 g of IV TXA during closure. Patients who had a history of deep venous thrombosis or pulmonary embolism, known thrombophilic disease, cardiac stent, ischemic stroke within 1 year, or seizures were given only 1 g of topical TXA during closure. Between March 1, 2018, and May 31, 2018, IV TXA was not used, and all patients received 1 g of topical TXA immediately before closure.
Anemia was defined as a hemoglobin value of less than 11.9 g/dL for women and less than 13.9 g/dL for men, based on the definition of anemia in the electronic medical record at the authors' institution. The decision to transfuse packed red blood cells was made at the discretion of hospitalists who consulted on every patient. Transfusion was considered for patients who had a hemoglobin value of less than 8 g/dL, according to the American Association of Blood Banks guidelines for orthopedic surgery.14 However, the authors' institution does not use a transfusion “trigger” based on hemoglobin values alone. Rather, the decision to transfuse was based on both laboratory values and symptoms of acute blood loss anemia, including but not limited to tachycardia, hypotension, orthostastic hypotension, lightheadedness, and shortness of breath.
The authors' institution charges $122.18 for a CBC without differential. A CBC with differential is charged at $203.63. Isolated tests for hemoglobin and hematocrit are charged at $84.64 and $94.62, respectively, for a cumulative charge for the 2 tests of $179.26. At the authors' hospital, either a CBC or an H&H is routinely ordered on postoperative days. Conservatively, the authors chose to use the lowest charge of $122.18 (CBC without differential) for the financial impact analysis.
Cost per intervention was calculated by dividing the cumulative cost of CBC testing by the number of interventions related to symptomatic anemia or abnormal laboratory values. The total potential savings (based on hospital charges) for the authors' hospital with the elimination of routine laboratory testing was calculated. The incidence of 90-day readmissions and postdischarge emergency department visits related to abnormal postoperative CBC values was examined as well.
Descriptive statistics were used for patient characteristics. Chi-square tests were used to determine whether there was a significant difference in the transfusion rate among patients with (1) preoperative anemia compared with those without preoperative anemia and (2) American Society of Anesthesiologists (ASA) score of 3 compared with ASA score of less than 3. The ASA scores were determined by the attending anesthesiologist, and the score was abstracted from each patient's chart. Exact tests were used when cell frequencies were small. P<.05 was considered significant. An odds ratio was used to determine whether the ASA score was related to abnormal findings on postoperative CBC. A CI of 95% was used, and P<.05 was considered significant. The effect of the ASA score was examined to determine whether the patient's state of health affected the utility of obtaining postoperative CBC testing.
During the study period, 647 patients underwent primary unilateral TKA. Of these, 19 (2.94%) patients did not meet the inclusion criteria and therefore were excluded from analysis (15 patients [2.32%] had not undergone preoperative CBC, 3 [0.46%] had not undergone postoperative CBC, and 1 [0.15%] required a return to the operating room because of a fall and patellar tendon rupture and distal femur fracture during hospitalization). After exclusion, 628 consecutive patients were analyzed. Patient demographics are shown in Table 1. Of the 628 patients, 26 (4.14%) had anemia preoperatively. A total of 390 patients (62.10%) had anemia postoperatively. An ASA score of 3 or greater was not associated with a greater risk of postoperative anemia (odds ratio [OR], 1.22; 95% CI, 0.85–1.74; P=.28). Mean hemoglobin values and mean decreases from preoperative measures to postoperative days 1 to 3 are shown in Table 2.
The 1 patient (0.16%) who was transfused 2 units of packed red blood cells was a 52-year-old woman with an ASA score of 3. She had chronic anemia as a result of systemic lupus erythematous. Her preoperative hemoglobin and hematocrit values were 11.4 g/dL and 37.7%, respectively. On postoperative day 3, she had tachycardia that did not respond to fluid resuscitation. Her hemoglobin and hematocrit values decreased to 8.2 g/dL and 26.5%, respectively. After workup excluded other possible causes of tachycardia, based on her symptoms, a decision was made to transfuse 2 units of packed red blood cells, resolving the tachycardia.
Because 26 of the 628 patients (4.14%) had preoperative anemia, the incidence of transfusion for patients with preoperative anemia was 1 of 26 (3.84%), which was significantly higher than that for patients without preoperative anemia (0%; P=.04). No difference was found in the transfusion rate for patients with an ASA score of 3 compared with patients with an ASA score of less than 3 (P=.14).
Within 90 days of surgery, no patients were readmitted for complications related to abnormal hemoglobin or hematocrit values. Although 31 patients were readmitted (mostly for manipulation under anesthesia) and 35 patients were seen in the emergency department for other reasons, none of them received an intervention because of low hematocrit or hemoglobin values.
A total of 956 routine CBC tests were obtained, at a conservative cost of $122.18 per CBC, for a total charge estimate of $116,804.08. If postoperative CBC testing was performed only for patients with preoperative anemia, then the total charges would be estimated at $3176.68 (or 2.71% of realized charges). The total charge per intervention in this study was $116,804.08. If postoperative CBC testing was performed only for patients with preoperative anemia, then the total charge per intervention would be estimated at $3176.68, suggesting a 97% cost reduction compared with current standard practice.
Because of the current trends of increasing health care costs and decreasing hospital reimbursements, strategies to minimize costs while maintaining patient safety are of paramount importance. There is renewed interest in challenging practices that are routinely used but have poor clinical utility. The current study challenges the utility of obtaining routine postoperative CBC testing after primary TKA and elucidates significant potential cost savings.
On the basis of their findings, the authors suggest that obtaining routine CBC or H&H testing for healthy patients undergoing primary TKA who receive TXA and are not anemic preoperatively is unnecessary from both a medical and an economic standpoint. Because none of the patients in this study were transfused based on H&H values alone and no patients without preoperative anemia required transfusion, routine postoperative CBC testing is recommended only if a patient who was not anemic preoperatively has symptoms of anemia postoperatively. On the other hand, patients who have baseline anemia preoperatively should be monitored closely for symptoms and may benefit from routine postoperative CBC testing (Figure 1) because 1 of 26 (3.84%) of the patients in this study who had anemia preoperatively required transfusion.
Algorithm to determine the need for a complete blood count (CBC) after primary total knee arthroplasty (TKA). Abbreviations: Hgb, hemoglobin; TXA, tranexamic acid.
In terms of patient safety, of these 628 patients, only 1 required postoperative blood transfusion, and the decision to transfuse was based primarily on symptoms (tachycardia) rather than a strict cutoff for hemoglobin or hematocrit value. Because this patient was anemic preoperatively as a result of chronic disease, according to the authors' protocol (Figure 1), postoperative laboratory tests would have been ordered routinely regardless, maximizing patient safety while limiting costs.
During the study period, the total hospital cost for obtaining routine CBC tests was $116,804.08. If the authors' hospital had limited obtaining CBC tests to patients with preoperative anemia, as proposed in Figure 1, $113,627.40 would have been saved, reflecting cost savings of greater than 97%. Moreover, during the time frame of this study, 2562 primary TKA procedures were performed by all surgeons at the authors' institution. A conservative estimate of CBC charges for all of these patients is, at a minimum, $313,025.16, based on calculating the cost of obtaining CBC testing on postoperative day 1 only. Because many of these patients underwent CBC testing on multiple postoperative days, that figure is likely higher, reflecting a substantial cost savings for the health care system.
Greco et al11 examined the utility of obtaining routine postoperative hemoglobin values after primary THA and TKA. All patients received TXA. They reported a similar rate of transfusion for patients undergoing TKA (0.27% [2 of 731 TKA procedures] vs 0.16% [1 of 628 TKA procedures] in the current study). They similarly concluded that patients receiving blood-conserving perioperative therapy, such as TXA, do not need postoperative CBC testing unless they have preoperative anemia. They did not examine the financial ramifications. Shaner et al12 concluded that routine postoperative laboratory testing is unnecessary for all patients after unicompartmental knee arthroplasty and showed potential cost savings if laboratory tests are not performed routinely. Kildow et al13 reported on the utility of CBC testing after THA and concluded that it was unnecessary for patients without preoperative anemia who received TXA. In their retrospective review of 352 patients, 54 (15.34%) of patients were transfused. However, not all of the patients received TXA. In addition, their study was performed at an academic institution, which is likely to have a more diverse patient population than the current study, which was performed at a nonteaching private practice institution.
This study had several limitations. First, it was performed at a private, nonacademic, orthopedic-only, inpatient hospital. This institution has a strict cutoff value for ASA score of 3 and body mass index of 50 kg/m2 and excludes patients who have severe systematic comorbidities. For that reason, the findings may not be generalizable to other institutions that have patient populations with greater demographic diversity and comorbidities. Further studies should be performed with a more generalizable patient population. Additionally, because of the retrospective nature of this study, the authors could obtain information on read-mission and emergency department visits only for their 17-hospital regional health care system. However, patients may have presented to an outside institution without their knowledge. Also, although ASA scores were abstracted from anesthesiology records, there is some inherent subjectivity with this score, and interobserver reliability of the ASA score among anesthesiologists has been shown to be moderate.15 For cost calculations, the authors used their hospital diagnosis-related group charges for CBC, which may have overestimated or underestimated the true cost to the hospital.
This study describes a potential way to reduce costs for patients undergoing primary TKA. Based on the results, healthy patients undergoing primary TKA with TXA do not require postoperative CBC testing, with potential cost savings of greater than 97% compared with the current routine practice of obtaining postoperative laboratory tests for every patient undergoing TKA. For patients with preoperative anemia, routine laboratory tests should be performed postoperatively to help to identify those who are at risk for requiring a blood transfusion, with a minimal effect on cost to the health care system. The potential health care savings associated with eliminating routine postoperative CBC testing are substantial and should be considered by orthopedic surgeons.
- Baker PN, van der Meulen JH, Lewsey J, Gregg PJNational Joint Registry for England and Wales. Data from the National Joint Registry for England and Wales: the role of pain and function in determining patient satisfaction after total knee replacement. J Bone Joint Surg Br. 2007;89(7):893–900. doi:10.1302/0301-620X.89B7.19091 [CrossRef] PMID:17673581
- Van Onsem S, Van Der Straeten C, Arnout N, Deprez P, Van Damme G, Victor J. A new prediction model for patient satisfaction after total knee arthroplasty. J Arthroplasty. 2016;31(12):2660–2667. doi:10.1016/j.arth.2016.06.004 [CrossRef]
- NIS overview. Accessed June 2019. https://www.hcup-us.ahrq.gov/nisoverview.jsp
- Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89(4):780–785. doi:10.2106/00004623-200704000-00012 [CrossRef] PMID:17403800
- One procedure that costs Medicare $6.6 billion. Accessed June 2019. https://www.cbsnews.com/news/one-procedure-that-cost-medicare-6-6-billion
- Kelly MP, Calkins TE, Culvern C, Kogan M, Della Valle CJ. Inpatient versus outpatient hip and knee arthroplasty: which has higher patient satisfaction?J Arthroplasty.2018;33(11):3402–3406. doi:10.1016/j.arth.2018.07.025 [CrossRef] PMID:30143333
- Backstein D, Thiagarajah S, Halawi MJ, Mont MA. Outpatient total knee arthroplasty: the new reality and how can it be achieved?J Arthroplasty. 2018;33(12):3595–3598. doi:10.1016/j.arth.2018.09.042 [CrossRef] PMID:30318253
- Wind TC, Barfield WR, Moskal JT. The effect of tranexamic acid on blood loss and transfusion rate in primary total knee arthroplasty. J Arthroplasty. 2013;28(7):1080–1083. doi:10.1016/j.arth.2012.11.016 [CrossRef] PMID:23541868
- Patel JN, Spanyer JM, Smith LS, Huang J, Yakkanti MR, Malkani AL. Comparison of intravenous versus topical tranexamic acid in total knee arthroplasty: a prospective randomized study. J Arthroplasty. 2014;29(8):1528–1531. doi:10.1016/j.arth.2014.03.011 [CrossRef] PMID:24768543
- Bagsby DT, Samujh CA, Vissing JL, Empson JA, Pomeroy DL, Malkani AL. Tranexamic acid decreases incidence of blood transfusion in simultaneous bilateral total knee arthroplasty. J Arthroplasty. 2015;30(12):2106–2109. doi:10.1016/j.arth.2015.06.040 [CrossRef] PMID:26235522
- Greco NJ, Manocchio AG, Lombardi AV, Gao SL, Adams J, Berend KR. Should postoperative haemoglobin and potassium levels be checked routinely following blood-conserving primary total joint arthroplasty?Bone Joint J. 2019;101-B(1 suppl A):25–31. doi:10.1302/0301-620X.101B1.BJJ-2018-0554.R1 [CrossRef] PMID:30648494
- Shaner JL, Karim AR, Casper DS, Ball CJ, Padegimas EM, Lonner JH. Routine postoperative laboratory tests are unnecessary after partial knee arthroplasty. J Arthroplasty. 2016;31(12):2764–2767. doi:10.1016/j.arth.2016.05.052 [CrossRef] PMID:27449718
- Kildow BJ, Howell EP, Karas V, et al. When should complete blood count tests be performed in primary total hip arthroplasty patients?J Arthroplasty.2018;33(10):3211–3214. doi:10.1016/j.arth.2018.05.030 [CrossRef] PMID:29908797
- Carson JL, Guyatt G, Heddle NM, et al. Clinical practice guidelines from the AABB: red blood cell transfusion thresholds and storage. JAMA. 2016;316(19):2025–2035. doi:10.1001/jama.2016.9185 [CrossRef] PMID:27732721
- Sankar A, Johnson SR, Beattie WS, Tait G, Wijeysundera DN. Reliability of the American Society of Anesthesiologists physical status scale in clinical practice. Br J Anaesth. 2014;113(3):424–432. doi:10.1093/bja/aeu100 [CrossRef] PMID:24727705
|Age at total knee arthroplasty, y|
| Mean (range)||67 (30–93)|
| Mean (range)||2.27 (0–3)|
|Body mass index, kg/m2|
| Mean (range)||30.76 (16.60–49.07)|
|Length of stay, d|
| Mean (range)||1.50 (1–5)|
| Left||304 (48.41%)|
| Right||324 (51.59%)|
| Female||411 (65.40%)|
| Male||217 (35.60%)|
|Preoperative hemoglobin, g/dL|
|Postoperative day 1 hemoglobin, g/dL (n=628)|
|Delta hemoglobin preoperative to postoperative day 1||1.8±0.9|
|Postoperative day 2 hemoglobin, g/dL (n=265)|
|Delta hemoglobin postoperative day 1 to postoperative day 2||0.3±0.8|
|Postoperative day 3 hemoglobin, g/dL (n=45)|
|Delta hemoglobin postoperative day 2 to postoperative day 3||2.6±2.9|