1 Introduction

Total knee arthroplasty (TKA) has transformed treatment for end stage osteoarthritis of the knee and is currently the most common joint replacement surgery in the United States. Although there have been significant advancements in implant designs, patient selection, and less invasive surgical techniques, it has been reported that around 20 % of patients remain dissatisfied following primary TKA.¹ Studies have shown that unlike the hip, the knee is rarely categorized as a forgotten joint following replacement, with associated symptoms of stiffness, pain, effusion, and discomfort commonly reported after TKA.^2,3

When evaluating the success after TKA, many surgeons utilize knee flexion range of motion (ROM) as the gold standard to assess short and long-term patient outcomes; 90° of knee flexion is considered to be the baseline target ROM following TKA, given that this range permits functional movement.⁴ Poor outcomes following TKA are usually characterized by a ROM short of 90° with associated arthrofibrosis, significantly lowering patient quality of life after TKA, and is the primary surgical indication for revision in about 4.5 % of TKAs annually.⁵ The first two weeks after TKA are considered a critical period for preserving knee ROM.^6,7 In fact, in this study by Wang et al., a group of patients received outpatient physical therapy (PT) for 2 weeks and another group receiver home exercise programs (HEP) for 2 weeks with the latter group undergoing outpatient PT at the 2 weeks follow-up if they had a knee flexion less than 90° or at the patient's request.⁷

Historically, formal outpatient PT was considered an essential component of the postoperative protocol after TKA, but in recent years, its necessity has been questioned due to recent studies demonstrating comparable clinical outcomes of patients utilizing self-directed HEP in lieu of formal PT.^7–9 Some studies have recently argued that although formal physical therapy is effective, it is often cost intensive, and could be an unnecessary burden to both the patient and the healthcare system.¹⁰ The results from these initial studies are worth considering, given that post-hospital rehabilitative care is known to be the largest driver in the variation in Medicare spending.¹¹ In addition to the financial burden stemming from formal PT, the issue of patient accessibility also deserves equal consideration. For some patients, physically attending a formal PT session is an added stress, given that many must rely on others for transportation in the weeks following surgery. Recent trends of therapist shortages, especially in medically underserved areas, only compound this issue.¹²

Initial studies demonstrating comparable clinical outcomes between patients prescribed formal PT vs HEP are promising, and more data is needed to consider HEP as an integral part of the postoperative protocol following primary TKA. The aim of this study is to investigate the impact of PT on ROM after TKA, with a null hypothesis that there is no difference between formal PT and HEP.

2 Material and methods

This is a randomized prospective study where 341 patients were prospectively randomized using computer generated simple randomization into formal PT or HEP after undergoing elective TKA between January 2017 to December 2018. Our practice's institutional review board approved the trial protocol (405019), and informed consent was provided by each patient enrolled in the study. A power analysis was done to identify the required sample size. With a significance level being 0.05, a clinically meaningful difference considered as 5°, and considering 10 % of missing values, the sample size required would be 220, 110 in each cohort.

In accordance with our institutional standard of care, all patients who underwent TKA during this time frame were provided with a home exercise program consistent with the American Association of Hip and Knee Surgeons (AAHKS) Home Therapy Exercises After Total Knee Replacement guide at time of discharge [Fig. 1].¹³ At their 2-week post-operative appointment for ROM and wound check, patients with 90° or greater of flexion were randomized to either:(1)No physical therapy with continuation of their home exercise program (HEP) (experimental),(2)Initiation of formal physical therapy (control).

Close

Fig. 1 The American Association of Hip and Knee Surgeons (AAHKS) Home Therapy Exercises After Total Knee Replacement guide.

Export to PPT

Patients who presented with less than 90° of flexion at their 2-week post-operative appointment were excluded from the study. Patients who were prescribed HEP were taught and provided a printout of the American Association of Hip and Knee Surgeons (AAHKS) Home Therapy Exercises After Total Knee Replacement guide [Fig. 1].¹³

Demographics, body mass index (BMI), and comorbidities (Charlson comorbidity index (CCI) were compared between the two groups in addition to knee ROM which was assessed at 6 weeks and 3 months. Patients were followed for a year until reaching maximum medical improvement (MMI). Emergency department visits for knee swelling, calf swelling, and medical admissions were recorded.

2.1

2.1 Characteristics

Of the 341 patients who underwent TKA, 301 (88.27 %) achieved 90° of flexion by their 2-week post-op visit. The 40 patients who achieved less than 90° of flexion at 2 week post-operatively visit were deemed high risk for stiffness and excluded from the study. The remaining 301 patients were offered participation in the trial with 63 of those patients opting for formal PT, and thus removed from the study. The remaining 238 patients were randomized to either PT or HEP, with 119 patients in each cohort. Twenty-one patients crossed over from no-PT to PT, and 7 patients in the PT group crossed over to the no-PT group [Fig. 2]. The mean age of the whole cohort was 67.3 years with 48.4 % being females, average BMI of 32.4, and mean CCI being 1.7. There was no difference in the demographics, BMI, and CCI between the two cohorts (p > 0.05).

Close

Fig. 2 Flowchart summarizing the inclusion of the patients.

Export to PPT

3 Results

3.1

3.1 Outcomes

At 6 weeks, the HEP group had a 109° of flexion compared to 106° in the PT group with no statistically significant difference (p > 0.05). Similarly, at 3 months, no difference was seen in the degree of flexion between the HEP (118°) and the PT group (121°) [Fig. 3].

Close

Fig. 3 Flexion ROM at 6 weeks and 3 months post-operatively in PT and HEP groups.

Export to PPT

In the PT group, 4 visits to the emergency department were recorded for knee swelling and incisional redness, 2 with calf swelling. In the HEP group, 2 patients were admitted for calf swelling. No deep venous thrombosis events were noted or other post-operative complications including revision surgery, irrigation and debridement, or peri-prosthetic fracture in this time frame.

4 Discussion

Recent literature supports that a formal unsupervised activity program should be recommended to all arthroplasty patients and that in a subset of these patients, an unsupervised activity program may be as efficacious as supervised physical therapy.¹⁴ Additionally, the use of supervised telerehabilitation has become increasingly popular with studies showing equivalent efficacies compared with traditional PT.¹⁴ This leads us to investigate the benefit of formal PT versus HEP in regards to post-operative ROM. In our investigation, there was no statistically significant difference in 6-weeks and 3-months post-operative knee ROM. However, the superiority in ROM in the HEP group at 6 weeks may be attributed to the fact that patients on their own may be hard on themselves and doing more than what is required while patients randomized to PT are following the instructions of the therapist and doing exactly as they are asked. As for the superiority in ROM at 3 months in the PT group, it may be attributed to the therapist going slowly with the patient to avoid stiffness and pain which may be seen if the patients did intensive PT, especially if they were not guided by a physiotherapist such as the case of the HEP group which achieved higher flexion ROM at 6 weeks. We also investigated the potential financial implications in recommending formal PT to all our post-operative TKA patients. In collaboration with our physical therapy, we estimate that the average reimbursement for a physical therapy visit is in the $75 range (varies based on insurance coverage), and the average length of physical therapy treatment for a total knee is 2x/week for 6 weeks (12 visits). $75 x 12 visits = $900. Thus, in our study with 119 patients who did not undergo formal physical therapy, cost savings are estimated to be around $107,100.

At a productive arthroplasty facility, the transition from formal physical therapy for all patients to home exercises programs for select patients has enormous implications in cost and resource savings. It is important to recognize that the reimbursement of therapy depends on various factors such as patient insurance, time and resource intensity of session per the therapist, and how many total sessions. Thus, our calculations are basic estimates of costs per patient and true cost savings require further analysis. Additionally, a subset of our patients that did not achieve greater than 90° of knee flexion by 2 weeks were defaulted to formal therapy which may imply that formal therapy continues to be essential in certain patients. Wang et al. reported mean savings ranging between $1340 and $1893 per patient after establishing HEP in addition to no difference in post-operative ROM and complications between PT and HEP.⁷ Such results were as well reported by other studies.^15–19 Nevertheless, it will require further investigation to truly clarify the benefit of formal therapy versus unsupervised home exercises.

The future is quite interesting with the potential implementation of tele-monitoring and web-based exercise programs. In fact, Centers for Medicare Services (CMS) have now introduced formal reimbursement for remote physiologic monitoring that could include direct sensors measuring exercise, range of motion, and general patient mobility. Recent clinical trials with TKA patients in Canada showed similar clinical outcomes and patient satisfaction for patients who participated in telerehabilitation programs compared to traditional PT.²⁰ In addition, Summers et al. were able to show even better post-operative ROM (up until 3 months), better patient reported outcomes measures and lower rates of manipulation under anesthesia in the telerehabilitation group compared to formal PT.²¹ Although telerehabilitation addresses patient accessibility, it does not account for the increasing concerns of therapist shortages. Future research should consider investigating the current modalities of remote monitoring in orthopedics, with the purpose of analyzing therapy cost and effectiveness. There is appreciable variability in the available options for remote physical therapy programs, and it would be helpful to compare the latest wearable technologies.

Our study is not without limitations. The main limitation is the short-term follow-up in these patients. Further studies in the future may focus on larger patient samples with long term follow-up to support our findings. It would also be beneficial to analyze patients' factors such as body habitus, multiple comorbidities, or poor pre-operative mobility that may predispose them to complications such that formal physical therapy resources may be utilized most efficiently. Another limitation is the potential selection bias in selection patients with ≥90° ROM at 2 weeks. However, our ultimate goal was to support that a select group of patients that achieve early ROM may not require formal PT and have equivocal outcomes and ROM with HEP.

5 Conclusion

In our prospective randomized study, patients who did not participate in formal physical therapy achieved equivocal knee ROM compared to patients who underwent the standard of care formal PT protocol. Our findings suggest that not all patients require formal physical therapy which has significant implications regarding resource utilization, cost savings, and patient satisfaction. Thus, we recommend consideration of select, motivated patients who have achieved >90° of flexion at first postoperative visit for self-guided home exercise programs rather than formal PT.

CRediT authorship contribution statement

Jonathan Liu: Writing and Data collection. Drew Clippert: Writing and Data collection. Mohammad Daher: Writing and Data collection. Noah Gilreath: Writing and Data collection. John Milner: Writing and Data collection. Eric M. Cohen: Supervision. Valentin Antoci: Supervision.

Consent

Consent was obtained from the patients.

Ethics

The Institutional IRB approval was granted before conducting this research.

Funding

None.