RAPT, which has been shown in previous studies to predict the length of hospital stay for TKA, consists of 12 items, including age, sex, walking distance, use of walking aids, and presence or absence of home care 5. Similar to the previous study findings, factors such as age and Barthel index on admission were related to the duration of TKA hospitalization. Furthermore, novel factors like pain, difference in anesthetic methods (surgical and anesthetic information), and the presence or absence of comorbidities on admission were also identified.
First, regarding the presence or absence of chronic pain, it is not uncommon for pain around the knee originating from the lumbar spine or hip present before surgery to remain after TKA 12. In addition, we also believe that chronic postsurgical pain (CPSP) after TKA should be considered. CPSP has been reported to occur in approximately 20% of TKA 13, and pain two months after TKA is often relieved by the subsequent course of therapy, even though there is a possibility that the pain may persist for 3–6 months or more after TKA 14, 15. The Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), which allows patients to assess their physical health as a specific measure, has been widely adopted for patients undergoing prosthetic joint surgery. Previous studies have reported that the pain, function, and stiffness scales of WOMAC were significantly higher in the preoperative assessment of patients who developed CPSP after surgery 16. Since lower scales of function and stiffness may cause a decline in the Barthel index, it was inferred that chronic pain and a low Barthel index on admission before surgery might lead to the onset of CPSP after TKA, which in turn affects the length of hospital stay after surgery. Further, the studies reported that patients with a Barthel Index of < 19 had shorter hospital stays, which may be due to their early transfers, but in this study, only 0.1% of the patients had a Barthel Index of < 19, and the sample size was small, so further detailed analysis is needed. In addition, height, weight, and BMI were significantly associated with length of stay. Several studies have shown that maintaining an appropriate body weight lessened the burden on the knee joint, leading to a reduction in pain. Therefore, it is necessary to estimate the length of hospital stay considering the increased risk of pain that continues after surgery in patients who do not have adequate weight control, have chronic pain before surgery, or have a low Barthel index on admission. Likewise, differences in the initiation of rehabilitation therapy after surgery affected the length of hospital stay. Therefore, it is important to understand the status of the Barthel index at the time of admission and start rehabilitation early.
Second, the difference between the bone implant and anesthetic method had an impact on the length of stay. Although general and local anesthesia, including spinal subarachnoid anesthesia, are suitable for TKA,, local anesthesia has been reported to shorten the length of hospital stay compared to general anesthesia 17–19, which was consistent with the findings of the present study. General anesthesia has been previously reported to prolong the length of hospital stay in patients undergoing unilateral TKA 20, 21, consistent with our findings. However, contrastingly, some studies have shown that the method of anesthesia does not affect the length of hospital stay 22. Nevertheless, studies have shown that for patients undergoing TKA, there is always a difference in postoperative pain experienced after either epidural or spinal anesthesia, and this is one of the reasons for the differences in the length of hospital stay 23, 24. Although these previous studies are randomized controlled trials, we believe that they involve small sample sizes because they are single-center studies, and further analysis of the effect of anesthesia methods on the length of hospital stay is needed to validate the results. In addition, bone transplantation has been associated with the length of hospital stay in this study. This may be because patients who require bone transplants have poor knee conditions. Patients with bad knees may take longer to rehabilitate. Therefore, patients who require bone transplants should consider the possibility of longer hospital stays.
Third, atrial fibrillation, a comorbidity on admission, emerged as an associated factor for longer hospital stay in patients undergoing TKA. In a previous study, multivariate analysis of the length of hospital stay in patients with comorbid atrial fibrillation reported acute coronary syndrome, acute decompensated heart failure, infection, heart failure with reduced ejection fraction (HFrEF), heart failure with preserved ejection function (HFpEF), elevated N-terminal (NT)-pro hormone brain natriuretic peptide (NT-proBNP) levels, and elevated combination of hypertension, abnormal liver/renal function, stroke history, bleeding history or predisposition, labile international normalized ratio, old age and drug/alcohol usage (HAS-BLED) as predictors of longer hospital stay 25. In addition, frailty is common in patients with atrial fibrillation, which possibly impacts therapies and outcomes 26, 27. Therefore, we inferred that patients undergoing TKA had suboptimal outcomes due to comorbidities, which may affect their length of hospital stay.
Fourth, the association with the annual number of TKA surgeries in an institution. Although the number of annual TKAs and length of hospital stay have not been previously reported.However, an analysis of the relationship between the number of annual surgeries with respect to the outcome after pancreatic cancer surgery or the mid-term mortality after coronary artery bypass surgery showed that the higher the number of annual surgeries, the higher the outcome and the lower the mortality rate 28, 29. This suggests that the higher the number of surgeries per year for TKA, the lower the complication and the shorter the hospital stay after surgery. We believe that further analysis of this factor is required in the future to compare the number of surgeries performed annually between facilities with high and low volume of surgeries.
It is worth noting that accurately estimating the length of hospital stay for patients undergoing TKA can increase bed turnover in medical institutions and reduce patient wait times 30. It can also reduce the cost of treatment, which can lead to a reduction in the burden of health insurance, cost of patient care, and savings in medical and health resources. In addition, an accurate estimation of the length of hospital stay may also reduce the potential risk of infection by reducing the probability of cross-infection between patients and shortening the time of direct contact between patients and physicians 31. In this study, we analyzed patients’ backgrounds and facility information obtained from the DPC data and identified relevant factors that can influence the length of hospital stay. To our knowledge, we believe this is the first study that uses as much as 13 pieces of information on comorbidities on admission to estimate the length of hospital stays for patients undergoing TKA.
A notable strength of this study is the sample size used. Several medical institutions in Japan have adopted DPC, which allows for large amount of data for our analysis. Nevertheless, in future, we will further increase the sample size and perform similar analyses for different surgical procedures.
The challenges and limitations of this study were primarily based on retrospective analyses of published databases. Attempts to statistically correct for biases have failed to rule out unmeasured and residual confounders (e.g., use and duration of different drugs and intraoperative blood loss) that may have affected the length of hospital stay. In addition, the hospital stays of our patients were relatively long compared to that of Western countries, and caution must be taken when applying our findings to different cohorts of patients. Therefore, it is necessary to validate whether the influencing factors mentioned in this study hold in prospective observations.
In conclusion, among the associated factors that can influence the length of hospital stay for patients undergoing TKAs, age, height, weight, BMI, Barthel index, method of anesthesia, bone transplant, timing of postoperative rehabilitation, atrial fibrillation, chronic pain, and the hospital volume of treated TKA cases, are of primary importance and should be highly considered.