Why Systematic Reviews Can’t Guide Clinically Relevant MLD Practice for Post-Knee-Replacement Clients

Image from Unsplash By Jon Tyson

Einstein also said “Not everything that counts can be counted, and not everything that can be counted, counts.”

Both these quotes capture the central problem: when it comes to clinical practice, what matters most to real-world MLD therapists simply isn’t what’s being measured in “Level I evidence”.

To understand this disconnect we need to revisit what evidence based practice really means.

Evidence based therapy is a synthesis of research evidence, clinical experience and observation, in the context of the desired outcomes of the recipient.

Relying on Level 1 evidence alone is just not enough. Yet systematic reviews and meta-analyses are held up as the pinnacle of 'evidence' in nearly all allied health settings.

Case-in-point, two recent systematic reviews on MLD in post-op rehab for total knee replacement concluded that MLD adds no measurable benefit.

Yet, most MLD Therapists, and the clients themselves, will report exactly the opposite.

What the Systematic Reviews Actually Examine

Systematic reviews such as those conducted by Lu et al. (2024) [1] and Migliorini et al. (2023)[2] only include data from randomised controlled trials (RCTs) that meet rigid methodological criteria.

Knee replacement surgery, 2-days post-op

Migliorini 2023 retrieved only four RCTs involving 197 knee replacements, with an average follow-up of seven weeks, and concluded that MLD produced no significant differences in pain, range of motion, or limb circumference [2].

Lu 2024 expanded the pool to seven RCTs (285 patients) but again limited interventions to 2–6 days immediately after surgery. Their meta-analysis reported no meaningful advantage in knee flexion or extension, circumference, or pain scores [1].

Both papers recommend that MLD not be used for post-surgical rehabilitation, but in both cases the populations and protocols of the included studies bear little resemblance to the clients seen weeks or months later in private MLD clinics.

The Unreal Client: Who Gets Included

Every “high-level” trial included in those reviews enrolled uncomplicated, primary arthroplasty patients: usually in their sixties, medically stable, and discharged within ten days. Exclusion criteria routinely removed anyone with venous insufficiency, diabetes, cardiovascular disease, obesity, chronic oedema, or delayed wound healing [3, 4}.

In contrast, therapists in private practice typically treat:

• People months after hospital discharge,

• Individuals whose standard rehabilitation has failed,

• Clients with multiple comorbidities, often combined with pain, stiffness, and psychosocial distress.

By design, such clients could never be enrolled in the studies on which systematic reviews are based. Therefore, the “evidence” is systematically stripped of clinical complexity.

The Unreal Intervention: What Counts as MLD

None of the meta-analysed RCTs investigated regionally integrated MLD protocols as used by certified Dr Vodder Therapists. Treatment sessions in the RCTs lasted 20–30 minutes, confined to the operated limb, and delivered by hospital physiotherapists following abbreviated protocols for three to five days post-surgery [3, 5].

In contrast, Vodder protocols address whole-body fluid dynamics, treating the neck, trunk, and contralateral limb to restore lymphatic balance and autonomic tone. Its purpose in the late post-operative phase is not merely to reduce centimetres of swelling, but to relieve fibrotic tension, pain hypersensitivity, and secondary oedema arising from reduced mobility or venous compromise. These goals fall entirely outside the outcomes chosen by reviewers.

The Unreal Outcomes: What Gets Measured

Systematic reviews judge “effectiveness” through numeric endpoints such as:

• range of motion measured by goniometer,

• mid-patellar circumference,

• visual analogue pain scores.

Such endpoints fail to reflect the functional recovery and quality-of-life outcomes valued by patients and therapists. When broader indicators are reported, they tell a different story, and although several of the included studies did find some benefit in MLD, these secondary outcomes do not make it into the conclusion of the systematic reviews.

For example, an included RCT by Cihan et al. (2021) found that adding MLD to standard rehab significantly reduced pain, fear of moving (kinesiophobia), and improved quality-of-life scores at six weeks [6].

Studies that were excluded from the reviews also tell a more clinically relevant story, and Zhang et al. (2019), working with diabetic knee-arthroplasty patients, reported superior knee flexion, reduced swelling, and better functional scores in the MLD group through six months of follow-up [4].

Studies that do not qualify for inclusion in systematic reviews because they do not meet narrow “Level I” design criteria arguably offer more clinically relevant insights than the studies that are included.

The Unreal Timeline: Acute Versus Chronic Needs

Across all Level I trials, intervention ceased within the first postoperative week and follow-up rarely extended beyond three months [1, 3]. For MLD Therapists in private-practice, treatment often begins long after the initial inflammatory phase, on clients who present with:

• persistent stiffness or fibrotic oedema at three-to-twelve months,

• compensatory gait dysfunction,

• lymphatic overload from comorbid lymphoedema, venous disease or obesity,

• secondary pain syndromes or complex regional changes.

Such chronic presentations involve physiological mechanisms such as connective tissue fibrosis, autonomic dysregulation, and fascial restriction that are not captured by short-term hospital studies of acute swelling. Evidence derived from the acute phase simply cannot predict outcomes in these chronic or multi-morbid contexts.

The Real-World Evidence: Individualised and Iterative

From Absolutely Maybe @PLOS Blogs

Clinically relevant evidence, is individually relevant, integrating applicable research with therapist expertise and client circumstances (7).

The existing RCTs, while methodologically rigorous, are contextually irrelevant: they study the wrong patients, at the wrong time, with the wrong protocols, and they measure the wrong outcomes.

When viewed through the lens of evidence-based practice, rather than practice-based evidence, these limitations mean that the absence of evidence for MLD in the early postoperative period does not equal no benefit of MLD in rehabilitation after knee replacement.

Case-series and practice-based evidence, although methodologically lower in hierarchy, often demonstrate meaningful functional gains such as improved gait symmetry, reduced night pain, and enhanced self-care participation - outcomes that matter to clients.

Toward a Clinically Relevant Research Model

To inform our MLD practice, future research must reflect real clinical populations:

• include older adults with comorbidities and chronic postoperative swelling;

• allow individualised Vodder-based or combined manual approaches;

• measure outcomes such as quality of life, kinesiophobia, function, and client-perceived recovery over six-to-twelve months.

Until such designs appear, systematic reviews limited to hospital-phase RCTs can only answer a narrow question: “Does brief, limb-only MLD change knee circumference within six weeks of surgery?”—not “Can skilled lymphatic therapy help complex patients regain movement and comfort months later?”

Conclusion

Systematic reviews deliver statistically neat but clinically sterile conclusions. They aggregate minimal-dose interventions in homogeneous hospital populations and then generalise the null results to every form of MLD.

For therapists working in private practice with clients who have failed standard rehabilitation and who live with the messy reality of comorbid lymphatic, venous, and metabolic dysfunction, such evidence offers little guidance.

Clinically relevant evidence arises not from erasing complexity but from engaging with it -through observation, documentation, and person-centred reasoning. The message for MLD practitioners is clear: systematic reviews may summarize research, but they do not define your clinical reality.

References:

1. Lu, H., Shao, Q., Li, W., Li, F., Xiong, W., Li, K., & Feng, W. (2024). Effects of manual lymphatic drainage on total knee replacement: a systematic review and meta-analysis of randomized controlled trials. BMC Musculoskeletal Disorders, 25(1), 30. doi:10.1186/s12891-023-07153-8

2. Migliorini, F., Schäfer, L., Bertini, F. A., Memminger, M. K., Simeone, F., Giorgino, R., & Maffulli, N. (2023). Level I of evidence does not support manual lymphatic drainage for total knee arthroplasty: a meta-analysis. Scientific Reports, 13(1), 22024.

3. Pichonnaz, C., Bassin, J.-P., Lécureux, E., Christe, G., Currat, D., Aminian, K., & Jolles, B. M. (2016). Effect of manual lymphatic drainage after total knee arthroplasty: a randomized controlled trial. Archives of Physical Medicine and Rehabilitation, 97(5), 674-682.

4. Zhang, H., Yan, J., Lin, S., Wang, H., Xiao, L., She, G., . . . Zha, Z. (2019). Manual Lymphatic Drainage Therapy in the Knee Joint Functional Rehabilitation After TKA in Diabetic Knee Osteoarthritis Patients: A Randomized Clinical Trial. Journal of Surgery, 7(3), 50-56.

5. Ebert, J. R., Joss, B., Jardine, B., & Wood, D. J. (2013). Randomized Trial Investigating the Efficacy of Manual Lymphatic Drainage to Improve Early Outcome After Total Knee Arthroplasty. Archives of Physical Medicine and Rehabilitation, 94(11), 2103-2111. doi: 10.1016/j.apmr.2013.06.009

6. Cihan, E., Yıldırım, N. Ü., Bilge, O., Bakar, Y., & Doral, M. (2021). Outcomes with additional manual lymphatic drainage to rehabilitation protocol in primary total knee arthroplasty patients: preliminary clinical results. Süleyman Demirel Üniversitesi Sağlık Bilimleri Dergisi, 12(3), 319-329.

7. Douglass, J. (2022). Evidence-based practice or practice-based evidence? Why pursuing level 1 evidence is leaving lymphoedema behind. Journal of Lymphoedema, 17(No 1), 8-11.