Machine Learning using Preoperative Patient Factors Can Predict Duration of Surgery and Length of Stay for Total Knee Arthroplasty

AbstractThe purpose of this study was to develop a predictive model for length of stay (LOS) after total knee arthroplasty (TKA). Between 2013 and 2014, 1,826 patients who underwent TKA from a single Singapore center were enrolled in the study after qualification. Demographics of patients with normal and prolonged LOS were analyzed. The risk variables that could affect LOS were identified by univariate analysis. Predictive models for LOS after TKA by logistic regression or machine learning were constructed and compared. The univariate analysis showed that age, American Society of Anesthesiologist level, diabetes, ischemic heart disease, congestive heart failure, general anesthesia, and operation duration were risk factors that could affect LOS (p < 0.05). Comparing with logistic regression models, the machine learning model with all variables was the best model to predict LOS after TKA, of whose area of operator characteristic curve was 0.738. Machine learning algorithms improved the predictive performance of LOS prediction models for TKA patients.

Download Full-text

Diagnosis of depression and other patient factors impacts length of stay after total knee arthroplasty

Arthroplasty Today ◽

10.1016/j.artd.2019.11.010 ◽

2020 ◽

Vol 6 (1) ◽

pp. 77-80 ◽

Cited By ~ 1

Author(s):

Cheongeun Oh ◽

Heather Gold ◽

James Slover

Keyword(s):

Total Knee Arthroplasty ◽

Length Of Stay ◽

Knee Arthroplasty ◽

Patient Factors ◽

Total Knee

Download Full-text

Trends in Periprosthetic Knee Infection and Associated Costs: A Population-Based Study Using National Data

The Journal of Knee Surgery ◽

10.1055/s-0040-1701516 ◽

2020 ◽

Author(s):

Robert Brochin ◽

Jashvant Poeran ◽

Khushdeep S. Vig ◽

Aakash Keswani ◽

Nicole Zubizarreta ◽

...

Keyword(s):

Total Knee Arthroplasty ◽

Length Of Stay ◽

Knee Arthroplasty ◽

Trend Test ◽

Strong Impact ◽

Hospitalization Costs ◽

Teaching Status ◽

Increasing Trend ◽

Total Knee ◽

Increasing Trends

AbstractGiven increasing demand for primary knee arthroplasties, revision surgery is also expected to increase, with periprosthetic joint infection (PJI) a main driver of costs. Recent data on national trends is lacking. We aimed to assess trends in PJI in total knee arthroplasty revisions and hospitalization costs. From the National Inpatient Sample (2003–2016), we extracted data on total knee arthroplasty revisions (n = 782,449). We assessed trends in PJI prevalence and (inflation-adjusted) hospitalization costs (total as well as per-day costs) for all revisions and stratified by hospital teaching status (rural/urban by teaching status), hospital bed size (≤299, 300–499, and ≥500 beds), and hospital region (Northeast, Midwest, South, and West). The Cochran–Armitage trend test (PJI prevalence) and linear regression determined significance of trends. PJI prevalence overall was 25.5% (n = 199,818) with a minor increasing trend: 25.3% (n = 7,828) in 2003 to 28.9% (n = 19,275) in 2016; p < 0.0001. Median total hospitalization costs for PJI decreased slightly ($23,247 in 2003–$20,273 in 2016; p < 0.0001) while median per-day costs slightly increased ($3,452 in 2003–$3,727 in 2016; p < 0.0001), likely as a function of decreasing length of stay. With small differences between hospitals, the lowest and highest PJI prevalences were seen in small (≤299 beds; 22.9%) and urban teaching hospitals (27.3%), respectively. In stratification analyses, an increasing trend in PJI prevalence was particularly seen in larger (≥500 beds) hospitals (24.4% in 2003–30.7% in 2016; p < 0.0001), while a decreasing trend was seen in small-sized hospitals. Overall, PJI in knee arthroplasty revisions appears to be slightly increasing. Moreover, increasing trends in large hospitals and decreasing trends in small-sized hospitals suggest a shift in patients from small to large volume hospitals. Decreasing trends in total costs, alongside increasing trends in per-day costs, suggest a strong impact of length of stay trends and a more efficient approach to PJI over the years (in terms of shorter length of stay).

Download Full-text

Intravenous patient-controlled analgesia vs nurse administered oral oxycodone after total knee arthroplasty: a retrospective cohort study

Scandinavian Journal of Pain ◽

10.1515/sjpain-2020-0012 ◽

2020 ◽

Vol 0 (0) ◽

Author(s):

Katarina Lahtinen ◽

Elina Reponen ◽

Anne Vakkuri ◽

Riku Palanne ◽

Mikko Rantasalo ◽

...

Keyword(s):

Total Knee Arthroplasty ◽

Length Of Stay ◽

Knee Arthroplasty ◽

Pain Medication ◽

Secondary Outcome ◽

Control Group ◽

Patient Controlled Analgesia ◽

Total Knee Arthroplasty Patient ◽

Opioid Dose ◽

Total Knee

AbstractShort CommunicationsSevere post-operative pain is common after total knee arthroplasty. Patient-controlled analgesia is an alternative method of pain management, whereby a patient administers his or her own pain medication. Patients seem to prefer this method over nurse-administered analgesia. However, it remains unclear whether patients using patient-controlled analgesia devices use higher or lower doses of opioids compared to patients treated with oral opioids.Objectives and MethodsThis retrospective study examined 164 patients undergoing total knee arthroplasty. Post-operatively, 82 patients received oxycodone via intravenous patient-controlled analgesia devices, while the pain medication for 82 patients in the control group was administered by nurses. The main outcome measure was the consumption of intravenous opioid equivalents within 24 h after surgery. Secondary outcome measures were the use of anti-emetic drugs and the length of stay. Furthermore, we evaluated opioid-related adverse event reports.ResultsThe consumption of opioids during the first 24 h after surgery and the use of anti-emetic drugs were similar in both groups. The median opioid dose of intravenous morphine equivalents was 41.1 mg (interquartile range (IQR): 29.5–69.1 mg) in the patient-controlled analgesia group and 40.5 mg (IQR: 32.4–48.6 mg) in the control group, respectively. The median length of stay was 2 days (IQR: 2–3 days) in the patient-controlled analgesia group and 3 days (IQR: 2–3 days) in the control group (p=0.02). The use of anti-emetic drugs was similar in both groups.ConclusionsThe administration of oxycodone via intravenous patient-controlled analgesia devices does not lead to increased opioid or anti-emetic consumptions compared to nurse-administered pain medication after total knee arthroplasty. Patient-controlled analgesia might lead to shortened length of stay.

Download Full-text