Total knee replacement (TKR) has proven to be an effective treatment modality for end-stage osteoarthritis, relieving pain and improving patients’ functional abilities. With an ageing and increasingly obese population, the number of people with end-stage osteoarthritis requiring TKR is expected to rise. In North America, the number of TKRs performed per annum is expected to increase by 673% in 2030, compared to the 2005 figures (Kurtz et al. 2007).

Locally in Singapore General Hospital, we see an increase of 100 TKRs per annum, with over 2,000 performed in 2019.


Chronic debilitating diseases such as knee osteoarthritis often represent a significant healthcare cost.

Two of the most effective ways to reduce the increasing healthcare cost associated with rising demand for TKR are to shorten the length of hospital stay and to minimise perioperative complications (Lovald et al. 2014).

There has been a trend towards earlier and safer patient discharges following TKR in the last two decades. The average length of a hospital stay has decreased from nine to four days at most hospitals in Singapore.

In recent years, there has been a push for enhanced recovery after surgery (ERAS) for TKR patients.

In August 2019, Singapore General Hospital (SGH) became the first hospital in Asia to launch an ERAS TKR clinical pathway, with carefully-selected patients safely discharged on postoperative day one (POD1) under the care of the Department of Orthopaedic Surgery.

This is achieved without putting the patients at increased risk for perioperative complications and a higher 30-day readmission rate.


Learning Points
Important components for a successful ERAS TKR clinical pathway include:
1. Preoperative patient stratification and selecting suitable patients
2. Proper preoperative counselling, with adequate information available for patients and their families
3. Optimisation of chronic diseases and preoperative medications
4. Perioperative physiotherapy, including preoperative and on POD0
5. Perioperative pain management, including good surgical technique with minimal soft tissue dissections
6. Optimal perioperative blood management
7. Perioperative intravenous fluid and nausea management
8. Avoiding the use of postoperative drain and in-dwelling urinary catheter (IDC)

Our criteria for ERAS TKR clinical pathway include patients scheduled for unilateral TKR with good family support, especially during the immediate postoperative period.

They will be assessed by the physiotherapist preoperatively and must be deemed to have good rehabilitation potential.

Exclusion criteria include TKR performed for knee fracture, patients with poorly controlled (HbA1c >8.0%) diabetes mellitus, bleeding disorders, ASA greater than II, poorly controlled cardiac or pulmonary comorbidities, chronic opioid use, functional neurologic impairments predisposing the patient to poorer rehabilitation potential, reduced preoperative cognitive capacity and preoperative voiding difficulties (Meneghini et al. 2017; Kort et al. 2017; Kerkhoffs et al. 2012).

We recommend proper preoperative counselling, and involving family support in the care of the patient during the immediate postoperative period.

This can significantly reduce the need for transfer to a community hospital after TKR. A local study showed that additional time spent inpatient at subacute community hospitals for inpatient physiotherapy does not confer better functional outcomes or quality of life to the patients
in the long term, when compared to home discharge with outpatient physiotherapy (Chan et al. 2018).

Preoperative education is also effective at preventing and reducing anxiety for both the patient and their family members. Pamphlets containing information on TKR can be given to patients to bring home to read before the actual surgery date.

Opioids and diabetic medications are the two most significant preoperative medications that predict longer inpatient hospital stays, more complications and a higher readmission rate after TKR (Rozell et al. 2017; Zarling et al. 2017).

We recommend referral to a Pain Team specialist and weaning off opioids prior to TKR, as well as optimising glycaemic control to reduce the length of hospital stay and perioperative complication rate.

  • Smoking cessation for six to eight weeks prior to TKR decreases the risk of infection, haematoma formation and wound complications (Ng et al. 2013).
  • Those with a history of alcohol abuse should abstain for at least one month before surgery. They should also be screened for liver disease and malnutrition.

Perioperative physiotherapy is a critical component that influences the outcomes after TKR. It also plays an integral role in the patient’s length of hospital stay.

This begins with preoperative physiotherapy, which allows patients to familiarise themselves with the exercise regimes that are expected of them after surgery, especially gait training with crutches, quad sticks or walking frames. Our postoperative physiotherapy protocol includes:

POD0: Self-assisted passive knee range of motion (ROM) exercises, isometric quadriceps strengthening exercises, unassisted straight leg raise exercises, standing and ambulating with assistive device, continuous passive motion (CPM) for ROM as per pain tolerance and cryotherapy on the operated knee.

POD1: Repeat ROM and strengthening exercises, and begin stair climbing exercises.

Good surgical technique that minimises soft tissue dissections during surgery, and the use of periarticular injection during TKR, has revolutionised postoperative pain management.

Randomised controlled trials have shown that it is as effective as patient-controlled analgesia (PCA) and femoral block, while avoiding the side effects of intravenous morphine and the unwanted increased risk of fall with femoral block (Song et al. 2016; Carli et al. 2010).

We routinely use periarticular injection without PCA. Our preferred cocktail for periarticular injection is marcaine with adrenaline, ketorolac, morphine, shincort and vancomycin if the patient has no contraindication. We inject into the medial parapatellar muscles where the arthrotomy is performed, as well as the periosteum and synovium to prevent adhesions, making sure that we make multiple passes with the needle, covering a wide area in the knee.

Periarticular injection during surgery should be coupled with multimodal pain management involving various analgesics with different mechanisms of action for pain relief (Gaffney et al. 2017; Lamplot et al. 2014). These include paracetamol, etoricoxib and gabapentin. Hence, we are able to achieve a synergistic effect with these analgesics and use a lower dose of each of them.

The need for pack cell transfusion can delay physiotherapy on POD0 and POD1. Patients may suffer from symptoms of anaemia preventing them from participating in their physiotherapy.

Therefore, we are of the opinion that the perioperative transfusion rate should be less than 1% in routine TKR.

A local study done at SGH found that for patients 70 years old and above, a preoperative serum haemoglobin level of more than 12.4 g/dL has a 99.3% negative predictive value for the need for perioperative transfusion.

Similarly, for patients under 70 years old, a preoperative serum haemoglobin level of more than 12.1 g/dL has a 99.5% negative predictive value for perioperative transfusion (Yeh et al. 2016).

For TKR patients with preoperative haemoglobin levels below these thresholds of 12.4 g/dL and 12.1 g/Dl, preoperative oral iron replacement is an option.

For severely anaemic patients, preoperative intravenous iron replacement and/or erythropoietic agents can be considered.

We routinely give intraoperative tranexamic acid to reduce perioperative blood loss during TKR (Chen et al. 2016; Shin et al. 2017).

Other methods to reduce perioperative blood loss in a multimodal approach include the use of hypotensive anaesthesia, intraoperative cell saver, as well as regional anaesthesia over general anaesthesia especially for bilateral TKR patients. Regional anaesthesia is associated with lesser perioperative blood loss and a 49% reduction in 30-day complication rate (Zhu et al. 2015).

Patients undergoing TKR should be kept well-hydrated. Dehydration is associated with postoperative complications including urinary retention, acute kidney injury and postural hypotension, all of which will hinder postoperative physiotherapy and increase the length of hospital stay.

Postoperative nausea is a common side effect of general anaesthesia, especially in younger women. Pre-emptive antiemetic medications used at SGH include ondansetron and metoclopramide, which are routinely ordered perioperatively. These two medications have a synergistic effect when given together to treat nausea, due to their different mechanisms of action. Ondansetron acts via the central nervous system while metoclopramide acts by increasing gastric outflow.

Proponents of the use of a drain during TKR feel that it will reduce intra-articular haematoma formation and serous leakage at the wound site, thereby reducing postoperative swelling and infection rate.

However, a study done at SGH comparing 575 TKRs with the use of a drain versus 902 TKRs without a drain found that the use of drain does not reduce the 30-day complication and readmission rates.

Neither does it reduce the incidence of additional surgical procedure performed on the same knee within two years after TKR. Of note, the use of a drain is associated with greater perioperative total blood loss (Chen et al. 2016).

While the use of a drain was the standard practice for many surgeons a decade ago, the Department of Orthopaedic Surgery at SGH has since moved away from it for routine TKRs.

We believe that the drain will impede immediate postoperative physiotherapy. We have not encountered a significant increase in wound complications since we stopped using drain in our practice.

We also advise against the routine insertion of an indwelling urinary catheter (IDC) in the operating room before TKR. To minimise the risk of postoperative urinary retention, we recommend avoiding prolonged surgery duration, as well as early mobilisation after TKR and adequate fluid hydration perioperatively to avoid acute kidney injury.


Deng et al. performed a systematic review and meta-analysis of ERAS, including 25 studies looking at TKR and/or total hip replacement (THR) outcomes. They concluded that ERAS significantly reduced the mortality rate, blood transfusion rate, incidence of perioperative complications and length of hospital stay for patients undergoing TKR or THR. However, ERAS did not have a significant impact on postoperative ROM and 30-day readmission rate (Deng et al. 2018).


The patient was a 64-year-old female with worsening mechanical left knee pain for three years. Her pain was associated with worsening deformity of her left knee. Her symptoms had progressively limited her ability to ambulate over the past three years and she was mostly homebound. Her quality of life had declined significantly.

She also had hypertension, diabetes mellitus and osteoporosis on bisphosphonate treatment. Her main caregivers were her maid and husband. She was using mostly oral etoricoxib and topical ketoprofen gel to relieve her left knee pain.

On clinical examination, the ROM of her left knee was 15 to 110 degrees. There was medial and lateral joint line tenderness, and patella grind test was positive.


Her preoperative serum haemoglobin was 12.2 g/dL, while a preoperative radiograph revealed a 22-degree varus deformity of her left knee.

She fulfilled our criteria for ERAS TKR clinical pathway.

  1. Preoperatively, she had been reviewed by a physiotherapist who introduced her to ambulation using a quad stick (which she required immediately after TKR) and educated her on what exercises to do perioperatively.
  2. She was also referred to an endocrinologist specialist to optimise her glycaemic control prior to her surgery.
  3. She was then scheduled in the morning on the operating room list, to allow better control of her serum glucose levels postoperatively prior to discharge, and thereby minimising the risk of perioperative complications (Ng et al. 2013).


She had her left TKR performed under regional anaesthesia, with an 11 cm midline skin incision via a medial parapatellar approach (Khakha et al. 2014). Bone cuts were made using conventional instrumentation.

She received fixed bearing, cruciate-retaining and cemented implants. The decision had been made to add a cemented stem to the tibial base plate, in view of her osteoporosis and severe preoperative varus deformity of the knee.

For analgesia, we routinely give a periarticular cocktail injection consisting of marcaine with adrenaline, ketorolac, morphine, shincort and vancomycin if patient has no contraindication. We do not use PCA.

The subcutaneous layer and skin were closed with dissolvable Stratafix suture and reinforced with tissue glue Dermabond® Prineo®. No drain or IDC was used.

Intraoperatively, she was given 1L of intravenous saline drip. Postoperatively, another 1L of intravenous saline drip was given over the next 23 hours. Preemptive antiemetic medications were also started, including ondansetron 4 mg Q8H and metoclopramide 10 mg Q8H.


Postoperative serum haemoglobin checks are not routinely ordered, except for preoperative anaemic patients or patients presenting with anaemic symptoms postoperatively.

Postoperatively, she was attended to by a physiotherapist on POD0 afternoon after the effect of her regional anaesthesia had worn off, and again on POD1 morning before discharge.

Our patient met our hospital discharge criteria:

  • She was able to flex her left knee till 110 degrees at 18 hours after her TKR, as well as ambulate and climb stairs independently with the help of a quad stick.
  • She was reviewed again in an outpatient clinic by a physiotherapist at one week after her surgery.


  1. Two of the most effective ways to reduce healthcare cost are to:
    • Shorten the length of hospital stay
    • Minimise perioperative complications after TKR
  2. Important components for a successful ERAS TKR clinical pathway include:
    • The selection of suitable patients
    • Proper preoperative counselling and patient education
    • Optimisation of chronic diseases and preoperative medications
    • Perioperative physiotherapy
    • Good perioperative pain, blood, fluids and nausea management


1. Kurtz S, Ong K, Lau E, Mowat F, Halpern M (2007) Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 89(4):780-785
2. Lovald ST, Ong KL, Malkani AL, Lau EC, Schmier JK, Kurtz SM, Manley MT (2014) Complications, mortality, and costs for outpatient and shortstay total knee arthroplasty patients in comparison to standard-stay patients. J Arthroplasty 29(3):510-515
3. Meneghini RM, Ziemba-Davis M, Ishmael MK, Kuzma AL, Caccavallo P (2017) Safe selection of outpatient joint arthroplasty patients with medical risk stratification: the “outpatient arthroplasty risk assessment score”. J Arthroplasty 32(8):2325-2331
4. Kort NP, Bemelmans YF, van der Kuy PHM, Jansen J, Schotanus MGM (2017) Patient selection criteria for outpatient joint arthroplasty. Knee Surg Sports Traumatol Arthrosc 25(9):2668-2675
5. Kerkhoffs GM, Servien E, Dunn W, Dahm D, Bramer JA, Haverkamp D (2012) The influence of obesity on the complication rate and outcome of total knee arthroplasty: a meta-analysis and systematic literature review. J Bone Joint Surg Am 94(20):1839-1844
6. Chan HY, Sultana R, Yeo SJ, Chia SL, Pang HN, Lo NN (2018) Comparison of outcome measures from different pathways following total knee arthroplasty. Singapore Med J 59(9):476-486
7. Rozell JC, Courtney PM, Dattilo JR, Wu CH, Lee GC (2017) Preoperative opiate use independently predicts narcotic consumption and complications after total joint arthroplasty. J Arthroplasty 32:2658-2662
8. Zarling BJ, Sikora-Klak J, Bergum C, Markel DC (2017) How do preoperative medications influence outcomes after total joint arthroplasty? J Arthroplasty 32:S259-S262
9. Ng VY, Lustenberger D, Hoang K, Urchek R, Beal M, Calhoun JH, Glassman AH (2013) Preoperative risk stratification and risk reduction for total joint reconstruction: AAOS exhibit selection. J Bone Joint Surg Am 95(4). e191-15.
10. Song MH, Kim BH, Ahn SJ, Yoo SH, Kang SW, Kim YJ, Kim DH (2016) Peri-articular injections of local anaesthesia can replace patientcontrolled analgesia after total knee arthroplasty: a randomised controlled study. Int Orthop 40(2):295-299
11. Carli F, Clemente A, Asenjo JF, Kim DJ, Mistraletti G, Gomarasca M, Morabito A, Tanzer M (2010) Analgesia and functional outcome after total knee arthroplasty: periarticular infiltration vs continuous femoral nerve block. Br J Anaesth 105(2):185-195
12. Gaffney CJ, Pelt CE, Gililland JM, Peters CL (2017) Perioperative Pain Management in Hip and Knee Arthroplasty. Orthop Clin North Am 48(4):407-419
13. Lamplot JD, Wagner ER, Manning DW (2014) Multimodal pain management in total knee arthroplasty: a prospective randomized controlled trial. J Arthroplasty 29(2):329-334
14. Yeh JZ, Chen JY, Bin Abd Razak HR, Loh BH, Hao Y, Yew AK, Chia SL, Lo NN, Yeo SJ (2016) Preoperative haemoglobin cut-off values for the prediction of post-operative transfusion in total knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 24(10):3293-3298
15. Chen JY, Chin PL, Moo IH, Pang HN, Tay DK, Chia SL, Lo NN, Yeo SJ (2016) Intravenous versus intra-articular tranexamic acid in total knee arthroplasty: A double-blinded randomised controlled noninferiority trial. Knee 23(1):152-156
16. Shin YS, Yoon JR, Lee HN, Park SH, Lee DH (2017) Intravenous versus topical tranexamic acid administration in primary total knee arthroplasty: a meta-analysis. Knee Surg Sports Traumatol Arthrosc 25(11):3585-3595
17. Zhu M, Chen JY, Tan YR, Yew AK, Chong HC, Chia SL, Lo NN, Yeo SJ (2015) Effects of anesthetic technique on blood loss and complications after simultaneous bilateral total knee arthroplasty. Arch Orthop Trauma Surg 135(4):565-571
18. Chen JY, Lee WC, Chan HY, Chang PC, Lo NN, Yeo SJ (2016) Drain use in total knee arthroplasty is neither associated with a greater transfusion rate nor a longer hospital stay. Int Orthop 40(12):2505-2509

Dr Jerry Chen currently works in the Department of Orthopaedic Surgery at Singapore General Hospital (SGH). He graduated with a Bachelor of Medicine, Bachelor of Surgery (MBBS) from Yong Loo Lin School of Medicine in 2010 and obtained his Fellowship of the Royal College of Surgeons (FRCS Orth) from Edinburgh in 2018. During his training years, he was awarded the Best Resident Award and Inspiring Resident Educator Award. He also published more than 45 scientific papers during his residency years and received over $500,000 of grant funding. Dr Jerry Chen’s clinical interests include hip and knee surgeries, minimally invasive surgery and ERAS clinical pathways.

Medical professionals who would like more information about this procedure, please contact Dr Chen at

GPs can call for appointments through the GP Referral Hotline at 6326 6060.