Abstract
Background
While advancements in surgery and reduced complication rates have made total knee arthroplasty (TKA) one of the most successful and cost-effective procedures in orthopaedic surgery, routine postoperative laboratory tests are still being ordered without evidence as to their necessity. With expansion of the bundled payment models, there may exist an opportunity to cut overall costs while maintaining quality of care by eliminating unnecessary interventions. The objective of this study was to examine the utility of routine postoperative laboratory tests in TKA.
Methods
A retrospective review of 319 TKAs performed at a single institution over a 2-year period was performed. The primary outcomes were the rates of acute blood loss anemia requiring transfusion, acute kidney injury (AKI), electrolyte abnormalities, and 90-day emergency department visits and readmissions. Multivariate logistic regression analysis was also performed to identify the risk factors associated with abnormal laboratory values.
Results
89 patients (27.9%) had abnormal postoperative laboratory results, of which 78% were exclusively due to electrolyte (sodium or potassium) abnormalities. The rates of AKI and blood transfusion were 3.8% and 1% respectively. Factors associated with electrolyte abnormalities were abnormal baseline electrolyte levels (p = 0.002 and = 0.006 for sodium and potassium respectively) and anemia (p = 0.049). Factors associated with blood transfusion were ASA score ≥3, preoperative anemia, and no tranexamic acid use. Factors associated with AKI were chronic kidney disease or having at least two of the following: age >65 years, BMI> 35, ASA score ≥3, diabetes, heart disease, and/or anemia. Laboratory results did not change the course of care in 305 of 319 patients (95.6%). There was no increased risk for 90-days ED visits or readmissions with abnormal laboratory values (p = 0.356).
Conclusion
With increasing pressure for cost containment in an era of bundled payment models, the very low rate of laboratory associated interventions suggest that routine postoperative laboratory tests is not justified. Obtaining laboratory after primary, unilateral TKA should be driven by patients’ risk factors.
Keywords: Total knee arthroplasty, Postoperative laboratory tests, Value-based medicine, Transfusion, Acute kidney injury, Electrolytes
1. Introduction
As bundled payment models become increasingly adapted, rising healthcare costs combined with fixed budgets are pressuring providers to deliver cost-effective care without comprising patient safety and outcomes.1,2 Cost-effectiveness is particularly important in joint arthroplasty as the demand continues to grow.3 Total knee arthroplasty (TKA) alone accounted for $3.5 billion in Medicare expenditure in 2013, which was Medicare's largest spending on a single procedure.4 Surgeons are currently utilizing various cost-saving strategies, such as improving efficacy in the operating room, optimizing clinical care pathway, minimizing hospital length of stay, and decreasing discharges to post-acute care facilities.5 While TKA is considered one of the most cost-effective surgeries in orthopaedic surgery,6, 7, 8 there still exists a need to explore other areas within TKA that can optimize its cost-effectiveness.
Routine postoperative laboratory tests continue to be ordered following TKA without evidence as to whether they are necessary. This is likely driven by historical practice of longer operative times, high blood loss, and increased complication rates.5 Today, with improved perioperative care pathways, routine laboratory tests may not be necessary. For example, the recent utilization of tranexamic acid (TXA) in orthopaedic surgery has been shown to significantly decrease intra-operative blood loss, transfusion rates, and overall hospital costs while maintaining low complication rates.9 There is currently limited data on the utility of routine postoperative laboratory testing after elective, primary TKA. Two recent studies from a single institution found that this practice was not indicated as only 1 in 3 basic metabolic panel (BMP) tests resulted in abnormal values requiring medical intervention10 and the risk of blood transfusion was very low with the use of tranexamic acid.11
The primary purpose of this study is to determine the incidence of abnormal postoperative laboratory studies following primary TKA and the frequency of abnormal labs resulting in interventions beyond the standard of care. A secondary aim of this study was to identify patient risk factors that would help predict those at higher risk for abnormal laboratory values.
2. Materials and methods
Institutional review board approval was obtained. This study was a retrospective chart review of all patients who underwent primary TKA at our institution between January 1, 2015 and July 15, 2017. Patients undergoing bilateral TKA, non-elective surgery, or major secondary procedures at the time of TKA were excluded. 396 procedures met these criteria. Of those, 77 patients were excluded for missing or incomplete preoperative laboratory values, leaving 319 patients available for analysis.
The following data were collected from the electronic medical records: age at time of surgery, sex, body mass index (BMI), American Society of Anesthesiologist Physical Status Classification (ASA), preoperative laboratory values, preoperative systolic blood pressure, intraoperative blood loss, tranexamic acid (TXA) use, postoperative laboratory values, any medical intervention directly related to abnormal laboratory values, and length of stay. Interventions were defined as medical or surgical treatments in direct response to abnormal laboratory values during the patients’ hospital stay (e.g. medical consults, transfusion, fluid bolus, electrolyte supplementation, etc.). Laboratory values collected were hemoglobin (Hgb), hematocrit (Hct), sodium (Na), potassium (K), and creatinine (Cr). Pre-operative laboratory values were compared to post-operative lab values. Normal laboratory values were defined as the following: Na (135–145 mEq/L),8 K (3.7–5.2 mEq/L),12 Cr (0.9–1.3 mg/dL in males and 0.6–1.1 mg/dL in females),13 and Hgb (13.5–17.5 g/dL in males and 12.0–15.5 g/dL in females).14 The primary outcomes were the rates of AKI (defined as an increase in baseline serum creatinine by at least 0.3 mg/dL or 26.4 mmol/L),15,16 Hgb of 8 g/dL or less, blood transfusion, electrolyte values outside of normal defined ranges,8,12 and 90-days emergency department (ED) visits and readmissions.
Two groups were compared: 1) patients with normal postoperative labs and 2) those with abnormal postoperative labs. Descriptive statistics including mean and standard deviation and frequency and proportion were calculated to characterize the study groups. An independent t-test was used to examine differences in continuous variables between the groups. The chi square or Fisher's exact test was used to compare differences in categorical variables. Multivariate logistic models were fit to identify factors associated with abnormal postoperative sodium and potassium values. The alpha level for all analyses was set at p < 0.05. All analyses were performed using Stata (StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC).
3. Results
319 patients were included in the analysis. The baseline characteristics between the study cohorts were similar except for higher rates abnormal Na levels (p = 0.007), abnormal K levels (p = 0.017), and lower use of tranexamic acid (p = 0.011) among patients who had abnormal postoperative laboratory values. Table 1 summarizes the characteristics of the study cohorts.
Table 1.
Characteristics of the study cohorts.
| Normal Postoperative Labs | Abnormal Postoperative Labs | P Value | |
|---|---|---|---|
| Preoperative Variables | |||
| N | 230 | 89 | |
| Age | 62.6 (±10.3) | 61.9 (±11.5) | 0.310 |
| Sex | 141F: 89 M | 52F: 37 M | 0.637 |
| BMI | 32.7 (±6.1) | 31.6 (±6.7) | 0.189 |
| ASA | 2.3 (±0.5) | 2.3 (±0.5) | 0.558 |
| Kidney Disease | 31 (14%) | 16 (18%) | 0.324 |
| Anemia | 54 (23%) | 30 (34%) | 0.063 |
| Diabetes | 40 (17%) | 16 (18%) | 0.902 |
| Abnormal Sodium (mEq/L) | 5 (2%) | 8 (9%) | 0.006 |
| Abnormal Potassium (mEq/L) | 11 (5%) | 11 (12%) | 0.017 |
| Intraoperative Variables | |||
| Estimated Blood Loss (mL) | 175 (±117) | 182 (±189) | 0.661 |
| Tranexamic Acid Use | 175 (76%) | 55 (62%) | 0.011 |
| Postoperative Variables | |||
| Change in Creatinine (mg/dL) | −0.02 (±0.13) | 0.01 (±0.22) | 0.129 |
| Change in Hemoglobin (g/dL) | 2.2 (±1.2) | 2.2 (±1.7) | 0.726 |
| Change in Sodium (mEq/L) | 1.6 (±2.6) | 3.5 (±3.7) | <0.001 |
| Change in Potassium (mEq/L) | 0.05 (±0.40) | 0.09 (±0.55) | 0.510 |
| Postoperative Events | |||
| Transfusion | 1 (0.4%) | 2 (2.2%) | 0.189 |
| Acute Kidney Injury | 0 (0.0%) | 12 (13.5%) | <0.001 |
| Length of Stay (days) | 3.5 (±0.8) | 3.7 (±1.5) | 0.031 |
| 90-day ED Visits or Readmissions | 3 (1.3%) | 3 (3.4%) | 0.354 |
BMI: body mass index; ASA: American Society of Anesthesiologists Physical classification system; ED: emergency department.
89 patients (27.9%) had abnormal postoperative laboratory results, of which 69/89 (78%) were exclusively due to electrolyte (sodium or potassium) abnormalities. 12 patients (3.7%) developed AKI and another 3 patients (1%) required transfusion with packed red blood cells for symptomatic blood loss anemia. Due to very low rates of blood transfusion and AKI, logistic regression analysis could not be performed to identify the risk factors associated with these outcomes. However, among the three patients who required blood transfusion, all had ASA score ≥3 and two had baseline anemia and did not receive TXA. Among patients who experienced AKI, 2/12 had baseline CKD. The remaining 9/12 patients could be divided into two groups: 1) those with ASA score ≥3 plus one of the following: preoperative anemia, diabetes, heart disease, BMI >35, or age >75 years; and 2) those with ASA score <3 plus two of the following: age >65 years, BMI ≥35, diabetes, or preoperative anemia. Table 2 summarizes the baseline characteristics of patients who experienced either an AKI or blood transfusion.
Table 2.
Characteristics of the patients who developed either symptomatic blood loss anemia requiring transfusion or acute kidney injury.
| Age (years) | Sex | Comorbidities | ASA Score | BMI | TXA Use | Baseline Hgb (g/dL) | Baseline Cr (mg/dL) | EBL (mL) | |
|---|---|---|---|---|---|---|---|---|---|
| Symptomatic Blood Loss Anemia Requiring Transfusion | |||||||||
| Patient 1 | 79 | F | Hypertension, CKD, gout, left bundle branch block, thalassemia trait, hyperthyroidism, hyperparathyroidism, anemia | 4 | 33 | No | 11.6 | 1.7 | 100 |
| Patient 2 | 53 | F | Depression, IBS, schizoaffective disorder, Kallmann syndrome | 3 | 28 | Yes | 12.8 | 0.77 | 300 |
| Patient 3 | 84 | F | CAD, history of MI, CVA, CHF, atrial fibrillation, GERD, pacemaker, anemia | 3 | 26 | No | 9.0 | 0.7 | 300 |
| Acute Kidney Injury | |||||||||
| Patient 1 | 70 | M | Diabetes mellitus, hyperlipidemia | 2 | 36 | Yes | 15.0 | 1.0 | 250 |
| Patient 2 | 55 | M | Hypertension, hyperlipidemia, anemia | 3 | 21 | Yes | 11.0 | 0.9 | 400 |
| Patient 3 | 69 | F | Hypertension, asthma, hyperlipidemia | 2 | 39 | Yes | 13.0 | 0.8 | 250 |
| Patient 4 | 88 | F | Hypertension, hyperlipidemia, anemia | 3 | 22 | Yes | 10.7 | 0.9 | 200 |
| Patient 5 | 78 | M | Hypertension, diabetes mellitus, PVD, history of thyroid cancer, history of prostate cancer, arrhythmia status post pacemaker placement | 3 | 36 | No | 13.4 | 1.1 | 250 |
| Patient 6 | 64 | M | Hypertension, BPH, history of MI, CAD, GERD | 3 | 29 | No | 15.2 | 0.9 | 136 |
| Patient 7 | 70 | F | Hypertension, history of breast cancer, hyperlipidemia, anemia | 2 | 35 | Yes | 9.8 | 0.9 | 100 |
| Patient 8 | 49 | M | Hypertension, alcohol dependence, tobacco dependence, CKD | 2 | 28 | 16.7 | 1.8 | 200 | |
| Patient 9 | 62 | M | Hypertension, hyperlipidemia | 3 | 42 | Yes | 14.5 | 0.9 | 250 |
| Patient 10 | 66 | M | None | 2 | 37 | Yes | 15.3 | 0.9 | 150 |
| Patient 11 | 64 | M | Hypertension, diabetes, anxiety, history of MI status post coronary stent placement | 3 | 29 | No | 13.5 | 0.8 | 200 |
| Patient 12 | 79 | F | Hypertension, CKD, gout, left bundle branch block, thalassemia trait, hyperthyroidism, hyperparathyroidism, anemia | 4 | 33 | No | 11.6 | 1.7 | 100 |
M: male; F: female, CKD: chronic kidney disease; IBS: irritable bowel syndrome; CAD: coronary artery disease; CVA: cerebrovascular accident; CHF: congestive heart failure; GERD: gastroesophageal reflux disease; PVD: peripheral vascular disease; BPH: benign prostatic hypertrophy; MI: myocardial infarction.
For abnormal postoperative Na levels, the most significant predictor was abnormal baseline Na levels (OR 5.79, p = 0.002). For abnormal postoperative K levels, there was a significant association with abnormal baseline K values (OR 4.47, p = 0.006) and anemia (OR 2.32, p = 0.049).
Laboratory results did not change the course of care in 305 of 319 patients (95.6%). Abnormal laboratory tests were associated with increased LOS (p = 0.031) but not 90-day ED visits or readmissions (p = 0.354).
4. Discussion
Today's healthcare climate is rapidly transitioning toward value-based bundled payment models that reward quality and efficiency.2 The Centers for Medicare and Medicaid Services (CMS), the largest payer for healthcare in the U.S., is actively leading this transition with a goal of linking least 50% of traditional fee-for-service payments to value-based payment models by the end of 2018.17,18 With joint replacement procedures already making up the largest proportion of inpatient surgical procedures for Medicare beneficiaries19 and with the demand expected to increase several folds by 2030,20 it is of vital importance to look for opportunities to curb unnecessary costs and resource utilization. This study examined the utility of routine postoperative laboratory tests as one potential opportunity to curb costs for TKA.
We found that despite 27.9% of patients undergoing primary, elective TKA had at least one abnormal postoperative laboratory finding, a laboratory associated intervention was extremely uncommon (4.4%). There rates of symptomatic blood loss anemia requiring transfusion was 3/319 (1%) and AKI was 12/319 (3.8%). The vast majority of abnormal laboratory tests (76%) were exclusively secondary to borderline abnormal electrolytes (sodium and/or potassium) and did not necessitate any interventions. There were no increased risk of ED visits or readmissions in the first 90 days after surgery related to abnormal laboratory values.
The risk factors for abnormal postoperative electrolyte levels were abnormal preoperative levels and anemia. The contribution of anemia to electrolytes imbalance may be explained by potential for increased crystalloid fluid administration to treat anemia-associated symptoms such as tachycardia, dizziness, and low blood pressure. The risk factors for blood transfusion were intuitive: ASA ≥3 and higher intraoperative blood loss especially in patients with baseline anemia or did not receive TXA. The risk factors for AKI were having at least two of the following: age >65 years, BMI> 35, ASA score >2, diabetes, heart disease, and/or anemia. These findings are consistent with previous studies. Park et al.21 retrospectively reviewed 4769 patients who underwent primary, unilateral TKA between 2000 and 2008 at a single institution. Male sex, higher BMI, hypertension, preoperative autologous blood donation, amount of intraoperative blood loss, Charlson Comorbidity Index >3, and lower preoperative Hgb levels were most predictive. Abar et al.13 retrospectively reviewed 1719 primary elective TJAs performed at a single institution between 2004 and 2015. The rate of AKI was 3.1% with the most significant risk factors being elevated preoperative creatinine levels, increased intraoperative blood loss, and higher ASA scores. Kildow et al.10 retrospectively reviewed 767 patients undergoing primary TJA and reported that the rates of postoperative sodium and potassium abnormalities were 41.3% and 25.2% respectively. Among the most significant risk factors for electrolyte abnormalities were female gender, CKD, advanced age, and diabetes.
The study should be interpreted in the context of some limitations. First, the data is derived from a single tertiary care academic center. Second, information on costs associated with laboratory tests was not reported as collecting this information was limited by variability in billing arrangements. Third, due to the retrospective nature of the study, it was not possible to determine whether the abnormal laboratory values, especially electrolytes abnormalities, were associated with any symptoms.
Last but not least, surgeons cannot just rely on laboratory tests. Clinical examination is paramount in the postoperative period and should take into account not just the surgical site but also indicators of overall physical health including vital signs, fluid intake/output, and symptoms. While excessive postsurgical bleeding may be readily appreciated on knee exam, it may not be so obvious following THA. The surgeon should also be aware of each patient's medical history and home medications and their potential adverse effects in the immediate postoperative period. For example, in our institution, we routinely hold oral hypoglycemics and non-beta blocker antihypertensives in the first 24 h until patients are able to resume adequate oral intake. This is done to prevent potential adverse events, such as hypotension-AKI and hypoglycemia. Furthermore, the use of nephrotoxic agents should be minimized or altogether avoided especially in patients with diabetes and/or chronic renal disease. A strategy aimed at preventing complications combined with good patient history and clinical examination cannot be overemphasized and it should be used to guide laboratory testing.
In conclusion, routine postoperative laboratory testing in modern day primary, unilateral TKA is not necessary in patients without risk factors. Electrolyte abnormalities rarely result in actionable information and most commonly occur in patients with preoperative electrolyte abnormalities and anemia. Elevated creatinine levels indicative of AKI are associated with CKD or having at least two of the following: age >65 years, BMI> 35, ASA score ≥3, diabetes, heart disease, and/or anemia. Acute hemoglobin levels drop requiring transfusion is associated with preoperative anemia, higher intraoperative blood loss, and lack of TXA use.
References
- 1.Iorio R., Clair A.J., Inneh I.A., Slover J.D., Bosco J.A., Zuckerman J.D. Early results of medicare's bundled payment initiative for a 90-day total joint arthroplasty episode of care. J Arthroplasty. 2016;31(2):343–350. doi: 10.1016/j.arth.2015.09.004. [DOI] [PubMed] [Google Scholar]
- 2.Centers for Medicare & Medicaid Services . 2018. Bundled Payments for Care Improvement (BPCI) Initiative: General Information.https://innovation.cms.gov/initiatives/bundled-payments/ Available at: [Google Scholar]
- 3.Van Manen M.D., Nace J., Mont M.A. Management of primary knee osteoarthritis and indications for total knee arthroplasty for general practitioners. J Am Osteopath Assoc. 2012;112(11):709–715. [PubMed] [Google Scholar]
- 4.Arthritis By The Numbers. Arthritis Foundation website. Available at: https://www.arthritis.org/Documents/Sections/About-Arthritis/arthritis-facts-stats-figures.pdf. Last accessed August 21, 2018. .
- 5.Aumiller W.D., Dollahite H.A. Advances in total knee arthroplasty. J Am Acad Physician Assistants. 2016;29(3):27–31. doi: 10.1097/01.JAA.0000480563.76217.ae. [DOI] [PubMed] [Google Scholar]
- 6.Waimann C.A., Fernandez-Mazarambroz R.J., Cantor S.B. Cost-effectiveness of total knee replacement: a prospective cohort study. Arthritis Care Res. 2014;66(4):592–599. doi: 10.1002/acr.22186. [DOI] [PubMed] [Google Scholar]
- 7.Losina E., Walensky R.P., Kessler C.L. Cost-effectiveness of total knee arthroplasty in the United States: patient risk and hospital volume. Arch Intern Med. 2009;169(12):1113–1121. doi: 10.1001/archinternmed.2009.136. discussion 1121-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kamaruzaman H., Kinghorn P., Oppong R. Cost-effectiveness of surgical interventions for the management of osteoarthritis: a systematic review of the literature. BMC Muscoskelet Disord. 2017;18(1):183. doi: 10.1186/s12891-017-1540-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lin Z.X., Woolf S.K. Safety, efficacy, and cost-effectiveness of tranexamic acid in orthopedic surgery. Orthopedics. 2016;39(2):119–130. doi: 10.3928/01477447-20160301-05. [DOI] [PubMed] [Google Scholar]
- 10.Kildow B.J., Karas V., Howell E. The utility of basic metabolic panel tests after total joint arthroplasty. J Arthroplasty. 2018;33(9):2752–2758. doi: 10.1016/j.arth.2018.05.003. [DOI] [PubMed] [Google Scholar]
- 11.Kildow B.J., Howell E.P., Karas V. When should complete blood count tests Be performed in primary total hip arthroplasty patients? J Arthroplasty. 2018 Oct;33(10):3211–3214. doi: 10.1016/j.arth.2018.05.030. [DOI] [PubMed] [Google Scholar]
- 12.Goodnough L.T., Schrier S.L. Evaluation and management of anemia in the elderly. Am J Hematol. 2014;89(1):88–96. doi: 10.1002/ajh.23598. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Abar O., Toossi N., Johanson N. Cost and determinants of acute kidney injury after elective primary total joint arthroplasty. Arthroplast Today. 2018;4(3):335–339. doi: 10.1016/j.artd.2018.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ahn J.H., Yang T.Y., Lee J.Y. Reduction osteotomy vs pie-crust technique as possible alternatives for medial Release in total knee arthroplasty and compared in a prospective randomized controlled trial. J Arthroplasty. 2016;31(7):1470–1475. doi: 10.1016/j.arth.2016.01.018. [DOI] [PubMed] [Google Scholar]
- 15.Lopes J.A., Jorge S. The RIFLE and AKIN classifications for acute kidney injury: a critical and comprehensive review. Clin Kidney J. 2013;6(1):8–14. doi: 10.1093/ckj/sfs160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Warth L.C., Noiseux N.O., Hogue M.H., Klaassen A.L., Liu S.S., Callaghan J.J. Risk of acute kidney injury after primary and revision total hip arthroplasty and total knee arthroplasty using a multimodal approach to perioperative pain control including ketorolac and celecoxib. J Arthroplasty. 2016;31(1):253–255. doi: 10.1016/j.arth.2015.08.012. [DOI] [PubMed] [Google Scholar]
- 17.Piccinin M.A., Sayeed Z., Kozlowski R., Bobba V., Knesek D., Frush T. Bundle payment for musculoskeletal care: current evidence (Part 2) Orthop Clin N Am. 2018;49(2):147–156. doi: 10.1016/j.ocl.2017.11.003. [DOI] [PubMed] [Google Scholar]
- 18.Centers for Medicare & Medicaid Services . 2016. Report to Congress December.https://innovation.cms.gov/Files/reports/rtc-2016.pdf Available at: [Google Scholar]
- 19.Centers for Medicare & Medicaid Services. Comprehensive Care for Joint Replacement Model. Available at: https://innovation.cms.gov/initiatives/cjr. Last accessed August 21, 2018.
- 20.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/JBJS.F.00222. [DOI] [PubMed] [Google Scholar]
- 21.Park J.H., Rasouli M.R., Mortazavi S.M., Tokarski A.T., Maltenfort M.G., Parvizi J. Predictors of perioperative blood loss in total joint arthroplasty. J Bone Joint Surg Am. 2013;95(19):1777–1783. doi: 10.2106/JBJS.L.01335. [DOI] [PubMed] [Google Scholar]