A network for students interested in evidence-based health care

Land‐based versus water‐based rehabilitation following total knee replacement: a critical appraisal of the evidence

Posted on June 3, 2019 by Lochlainn Murray

Evidence Reviews

This blog is a critical appraisal of the following randomized controlled trial: Land-based versus water-based rehabilitation following total knee replacement: a randomized, single-blind trial.


A Total Knee Replacement (TKR), is a surgical procedure where a damaged joint is replaced with an artificial joint (NHS, 2016). This surgery can be used to treat patients whose condition causes severe knee pain and stiffness. In a study carried out by Kurtz et al. (2007), it is estimated that the number of TKRs will increase by up to 7 times for arthritic patients by 2030. The main problems post-surgery are: decreased muscle strength, decreased range of movement (ROM), altered gait, and increased pain and swelling (Lin He et al 2014). Hydrotherapy is an under-utilised physiotherapy treatment post-TKR. There is limited evidence to show whether it is more effective than conventional physiotherapy.

How was the study carried out?

This study investigated the effects of a water-based physiotherapy program compared to land-based physiotherapy on patients following a TKR. 102 patients consented out of 216. They were screened against exclusion criteria, e.g. contracting a post-operative deep joint infection, bilateral joint surgery/further surgery within six months. Patients with a neurological condition affecting informed consent were excluded. 84 participants were excluded. 53 patients went into the water-based treatment group and 49 into the land-based group. Hydrotherapy consisted of two 1-hour sessions per week. Land-based treatment followed a conventional physiotherapy program. All patients exercised at their own intensity.  Both groups were led by the same physiotherapist. They were encouraged to follow a Home-Exercise-Program (HEP) but this was not monitored.

A random number generator assigned patients a number, and a blinded researcher drew sealed envelopes to randomly assign patients to either group. An independent blinded examiner measured the outcomes. The primary outcome measured was the 6-Minute-Walk-Test (6MWT) and the secondary outcomes measured were the Stair-Climbing-Power-test (SCP), the WOMAC scale, the Visual-Analogue-Scale (VAS), knee ROM and knee oedema. These were tested after 2, 8 and 26 weeks post surgery and measured against baseline results.


There were significant improvements for the hydrotherapy group, with the mean 6MWT increasing by 201m (P<0.000). There were no significant differences when compared with the land-based group, with 3m mean difference. Both groups made significant improvements in both primary and secondary outcome measures (P<0.005 for both) with confidence intervals of 95%.

Strengths and weaknesses

The CASP RCT Checklist was used to analyse this study (CASP 2018). RCT was an appropriate design for this study as randomization minimizes bias and comparison tests the effectiveness of intervention.

A strength of this study was patient randomization. This minimizes potential bias and increases internal validity. It was impossible to blind the physiotherapist as they were administering the treatment. This makes results vulnerable to bias. The outcome examiner was independent and could be blinded to the group allocation. This reduces rater bias. However, a weakness of this study is when the independent examiner was unavailable on a few occasions, the data became vulnerable to bias. The data was then collected by an un-blinded therapist.

Another strength was the outcome measures used. The 6MWT tests aerobic capacity and endurance. It has a high test-retest reliability so the results between the two groups can be reliably compared. Secondary outcome measures that tested both physical (SCP and knee ROM) and functional capacity (WOMAC scale) were used. Functional training of the muscles is more appropriate than strengthening alone. This makes rehab more meaningful for the patients and could tie in with their functional goals. These are objective measures. VAS can give a subjective measure to how the patient feels.

At base-line, there were no differences between groups except 37% of the land-based group had heart disease compared to 17% in water-based. Both groups had a similar mean age; hydrotherapy was 68.7 years and land-based was 67.8 years. There was a similar ratio of male-to-female between groups with 57% being female in both. This is similar to the general population. There was little difference in the mean BMI’s, with exception to 8% of the study who were morbidly obese. The more similar the groups, the greater chance any improvements recorded are due to intervention and not chance (further reading in this blog explains that concept very well). Treatment given was similar in frequency; 2 x a week and a duration of 1 hour. The intervention was the only variable changed.

The results from this study could be applied to the wider population due to wide inclusion and narrow exclusion criteria. A narrow exclusion criteria allows for patients with co-morbidities to participate, who may have been excluded otherwise. This makes the sample similar to the population. Only patients with co-morbidities that would directly influence results were excluded.

Five pieces of data were also missing, with 3 sets in the hydrotherapy group missing and 2 sets in the land-based. This was done on an intention-to-treat basis. 2 patients in the water-based group declined the 8-week assessment but were present for the final assessment. 1 patient died of cardiac issues after 4 weeks. 2 patients missed the final session from the  land-based group as one relocated and one was hospitalised. The data set is incomplete, decreasing reliability of results.

The sample size was another weakness. This sample was a relatively small size to apply to the general population. A small sample can allow anomalous results to have a big impact on results, whereas a larger sample size minimizes this risk and is more reliable.


This study found significant improvements on physical and functional capacity for TKR patients. However, there were no significant differences found between land-based and water-based treatment. Therefore, the study cannot say one method is superior to the other as more research is required on a larger scale.


Critical Appraisal Skills Programme (2018). CASP (Randomised Controlled Trial) Checklist. [online] Available at: URL. https://casp-uk.net/wp-content/uploads/2018/03/CASP-Randomised-Controlled-Trial-Checklist-2018_fillable_form.pdf Date Accessed: 01/12/18.

He  ML, Xiao  ZM, Lei  M, Li  TS, Wu  H, Liao  J. Continuous passive motion for preventing venous thromboembolism after total knee arthroplasty. Cochrane Database of Systematic Reviews 2014, Issue 7. Art. No.: CD008207. DOI: 10.1002/14651858.CD008207.pub3.

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 Surgery Am. 89(4):780–785. Date Accessed: 01/12/18.

NHS (2016). National Health Service Conditions- Total Knee Replacement. [online] [10/12/18]. Available from- www.nhs.uk/conditions/knee-replacement. Accessed: 01/12/18.


Lochlainn Murray

Physiotherapy student at Queen Margaret University, Edinburgh. View more posts from Lochlainn

Leave a Reply

Your email address will not be published. Required fields are marked *

Subscribe to our newsletter

You will receive our monthly newsletter and free access to Trip Premium.