A Posteromedial-Corner Cause of Knee Pain
- An OVERUSE tendinopathy of the DISTAL SEMIMEMBRANOSUS at its posteromedial tibial insertion: chronic POSTEROMEDIAL knee pain, tender over the insertion, worse on RESISTED KNEE FLEXION
- The key differential is a POSTERIOR-HORN MEDIAL MENISCUS tear; localise the tenderness and correlate with imaging
- NON-OPERATIVE first: relative rest, ECCENTRIC and progressive loading, correction of training load and biomechanics
- CORTICOSTEROID helps in the short term but the benefit REVERSES at intermediate and long term (41 RCTs, 2,672 participants), with a 0.1% tendon rupture rate; a diagnostic local-anaesthetic block is a separate decision
- SURGERY is rare and reserved for refractory cases; no series exists for the distal insertion, and in PROXIMAL hamstring tendinopathy 80 of 90 returned to sport at a mean of 5 months
- On the differential of the painful TKA: excising the un-released REFLECTED arm relieved the pain completely in 5 of 6
- “Semimembranosus tendinopathy = overuse tendinopathy at the distal SM insertion (posteromedial tibia) = chronic POSTEROMEDIAL knee pain; tender over the insertion, worse on RESISTED KNEE FLEXION; running/cutting athletes.
- “Differential (posteromedial corner): posterior-horn medial meniscus tear, pes anserine bursitis (more anteromedial), popliteal/Baker's cyst, posteromedial-corner/MCL pathology, medial OA - localise + image.
- “Ultrasound (and MRI) characterise it and exclude meniscal pathology - though no accuracy figure exists at this insertion. Management NON-OPERATIVE first (rest, eccentric loading, correct training load); surgery rare.
- “STEROID CAVEAT: across 41 RCTs / 2,672 participants, corticosteroid helped short-term but the effect REVERSED at 26 and 52 weeks, with 0.1% tendon rupture. Use it to open a rehabilitation window, not as treatment. A diagnostic local-anaesthetic block is a different thing.
- “Don't forget the PAINFUL TKA: posteromedial pain within a year, worse rising from a chair, good alignment, no infection - excising the un-released reflected SM arm cured 5 of 6.
Overview and Presentation
Semimembranosus tendinopathy is an overuse tendinopathy of the distal semimembranosus at or near its insertion on the posteromedial tibia. It is a recognised, though under-recognised, cause of chronic posteromedial knee pain. The tendon has a complex multi-armed distal insertion that also stabilises the posteromedial corner of the knee.
Presentation. The patient is typically a running, cutting or repetitive knee-flexion athlete with chronic, localised posteromedial pain and tenderness over the insertion. The pain is characteristically worse on resisted knee flexion and with activity. The diagnosis is clinical, supported by imaging.
Anatomy: The Five-Arm Insertion
The distal semimembranosus fans into five arms:
- Direct (tibial) arm - the main tendon, inserting into the tuberculum tendinis on the posteromedial tibia just below the joint line, deep and posterior to the superficial MCL. This is the arm at which the insertional tendinopathy characteristically occurs.
- Anterior (reflected) arm - turns anteriorly deep to the superficial MCL to insert on the medial tibia beneath it.
- Oblique popliteal ligament (OPL) arm - a broad capsular expansion running superolaterally across the posterior capsule to form the oblique popliteal ligament, reinforcing the posterior capsule against hyperextension.
- Posterior oblique ligament (POL) / capsular arm - an expansion blending into the posteromedial capsule and POL.
- Arm to the posterior horn of the medial meniscus (and popliteus fascia) - a slip attaching to the posterior horn of the medial meniscus.
Function. The semimembranosus flexes and internally rotates the knee and is the principal dynamic stabiliser of the posteromedial corner. Through its meniscal arm it retracts the posterior horn of the medial meniscus during flexion, protecting it from impingement. That is precisely why posteromedial-corner pain from the tendon and from a posterior-horn meniscal tear can be so hard to separate.
Imaging
Ultrasound and MRI. Ultrasound is a useful alternative or adjunct to MRI for assessing posteromedial-corner pathologies, semimembranosus tendinopathy included. MRI characterises the tendon and excludes a meniscal tear and other causes. No diagnostic-accuracy figure for either modality has been published at this insertion.
Differential Diagnosis
Posteromedial knee pain has several possible sources, and the semimembranosus insertion is only one of them:
- Posterior-horn medial meniscus tear, which can present with very similar posteromedial pain
- Pes anserine bursitis or tendinopathy, usually a little more anteromedial
- Popliteal (Baker's) cyst
- Posteromedial-corner or medial collateral pathology
- Referred medial-compartment osteoarthritis
Confirm the source first. Before attributing the pain to the tendon and committing to prolonged tendinopathy treatment, localise the tenderness precisely to the semimembranosus insertion, provoke the pain with resisted knee flexion, and correlate with imaging. Localisation and imaging together are what separate the neighbours.
Against a posterior-horn meniscal tear. The closest differential separates on five features:
- Semimembranosus tendinopathy
- Precisely over the insertion (posteromedial tibia, at and just below/behind the joint line)
- Posterior-horn medial meniscus tear
- Along the posteromedial joint line
- Semimembranosus tendinopathy
- Resisted knee flexion, which loads the tendon
- Posterior-horn medial meniscus tear
- Joint-line rotation tests: McMurray (deep flexion with external rotation and extension), Thessaly (weight-bearing rotation) and Apley grind
- Semimembranosus tendinopathy
- None
- Posterior-horn medial meniscus tear
- Catching, locking, giving-way and an effusion
- Semimembranosus tendinopathy
- Gradual overuse (running/cutting)
- Posterior-horn medial meniscus tear
- A twisting/loading event, or an insidious degenerate tear in older patients
- Semimembranosus tendinopathy
- Thickened, hypoechoic or high-signal degenerate tendon at the insertion on ultrasound/MRI
- Posterior-horn medial meniscus tear
- MRI meniscal signal that reaches the articular surface
The diagnostic block. A diagnostic image-guided local-anaesthetic injection to the semimembranosus insertion is a useful tie-breaker: pain relief confirms the tendon as the source.
Management
Non-operative first. The programme combines:
- Relative rest and activity modification
- Eccentric and progressive loading rehabilitation
- Correction of training load and biomechanics
- NSAIDs
- An image-guided injection for refractory cases (see the corticosteroid caveat below)
The pathology. Biopsy studies of hamstring tendinopathy show tendinosis, a degenerative rather than inflammatory process. That is the argument for load-based rehabilitation over anti-inflammatory-led treatment.
Surgery. Surgery is rarely required. It is reserved for refractory cases, after a thorough non-operative programme and once the diagnosis is secure, with the semimembranosus insertion confirmed as genuinely the pain source.
What to tell the patient. No surgical series exists for the distal insertion. In 103 cases of proximal hamstring tendinopathy, results were excellent or good in 92 and 80 of 90 returned to their previous sporting level, but it took a mean of 5 months (range 2-12). Counsel months, not weeks.
The Corticosteroid Injection: Short-Term Gain, Long-Term Loss
An image-guided corticosteroid injection is a reasonable option for a refractory insertional tendinopathy. It must be offered with the caveat that the best evidence in tendinopathy shows the benefit reverses over time.
The evidence. A systematic review of 41 randomised trials covering 2,672 participants found, consistently across high-quality trials, that corticosteroid injection reduced pain in the short term but that the effect was reversed at intermediate and long term. For lateral epicondylalgia the short-term effect was large (SMD 1.44, 95% CI 1.17-1.71), yet at 26 weeks and at 52 weeks no intervention was favoured over injection (SMD −0.40, 95% CI −0.67 to −0.14 at intermediate term; −0.31, −0.61 to −0.01 at long term). Patients injected did worse by a year than those left alone.
Rupture. Among 991 participants who received corticosteroid injections in trials reporting adverse events, one (0.1%) sustained a tendon rupture. It is rare, but it is the complication to mention when consenting, particularly for a load-bearing insertional tendon.
The site caveat. Effects varied by tendon site, and the same authors explicitly warn that the response to injection should not be generalised across sites; short-term efficacy in rotator-cuff tendinopathy, for instance, was not clear. No trial has studied injection at the semimembranosus insertion at all, so these figures are the best available guide, not site-specific evidence.
In practice. Load-based rehabilitation remains the mainstay. The injection buys a window of analgesia in which to rehabilitate rather than being a treatment in itself, and repeated injections into an insertional tendon are not supported.
A local-anaesthetic block that abolishes the pain is a diagnostic test, and a valuable one: it identifies the tendon as the source. Injecting corticosteroid for treatment is a separate decision, with a reversal of benefit by six to twelve months and a small rupture risk. Do not let a positive diagnostic block automatically become a therapeutic steroid injection.
The Painful Posteromedial Knee After Total Knee Replacement
Semimembranosus tendonitis belongs on the differential of the unexplained painful total knee arthroplasty. In a series drawn from more than 700 total knee arthroplasties performed over eight years, 6 patients developed posteromedial knee pain within a year of surgery.
The clinical picture. Pain increased on walking and especially on rising from a chair. All six were women, and all had good knee motion and well-aligned, well-fixed implants, so the arthroplasty itself did not explain the pain. There were no signs of low-grade infection, which must still be excluded first in any painful replaced knee.
Confirmation and treatment. Injection of local anaesthetic relieved the pain, briefly, in all six. The culprit was the reflected portion of the semimembranosus tendon, which had not been released during the arthroplasty; excision relieved the pain completely in 5 of 6, with improvement in the sixth.
Reading the denominator. Six cases from over 700 arthroplasties is roughly 1%. This is a retrospective single-centre series with no stated follow-up duration for the six, so it establishes the entity and a plausible remedy rather than a reliable incidence or success rate.
Mnemonics & Memory Aids
POSTEROMEDIAL
Hook:Posteromedial pain, Resisted flexion, Overuse, Meniscus differential, Exclude neighbours, Diagnose (US/MRI) + non-op first.
Clinical Decision Scenarios
Practise clinical reasoning and management decisions out loud
“A runner has chronic posteromedial knee pain, tender over the semimembranosus insertion and worse on resisted flexion. How do you approach it?”
What it is
- Overuse tendinopathy of the distal semimembranosus (posteromedial tibial insertion)
- Cause of chronic posteromedial-corner knee pain
- Running/cutting/repetitive-flexion athletes
Presentation & differential
- Localised posteromedial tenderness; worse on resisted knee flexion
- Differential: posterior-horn medial meniscus, pes anserine bursitis (more anteromedial)
- Popliteal/Baker's cyst; posteromedial corner/MCL; medial OA
Diagnosis & management
- Ultrasound (and MRI) characterise the tendon + exclude meniscal pathology (no accuracy figure published)
- Non-operative first: rest, eccentric/progressive loading; correct training load/biomechanics
- Steroid injection: 41 RCTs / 2,672 participants - short-term gain REVERSED at 26 and 52 weeks; 0.1% rupture
- Diagnostic local-anaesthetic block confirms the source; that is not the same decision as a steroid injection
- Surgery rare; proximal hamstring series - 80 of 90 back to sport, but mean 5 months (range 2-12)
The painful TKA scenario
- Posteromedial pain within a year of arthroplasty, worse walking and especially RISING FROM A CHAIR
- Good motion, well-aligned prosthesis, no evidence of infection (still exclude it first)
- Local-anaesthetic block relieves it; culprit is the un-released REFLECTED arm
- Excision relieved pain completely in 5 of 6 patients (from over 700 arthroplasties)
Evidence & Key Studies
Ultrasonography for posteromedial-corner pathologies including semimembranosus tendinopathy
- A NARRATIVE REVIEW, not a diagnostic-accuracy study - it recommends ultrasound but reports no sensitivity, specificity or yield figure for detecting semimembranosus tendinopathy, and none has been published.
- The posteromedial corner (PMC) is a common location for knee pain, with frequent causes including posterior medial meniscus pathology, articular cartilage damage, popliteal cysts and semimembranosus tendinopathy.
- While MRI is the gold standard, ultrasonography is a useful alternative for assessing the posteromedial corner, with particular value in evaluating semimembranosus tendinopathy.
- Accurate localisation and imaging assessment of the posteromedial corner are essential for appropriate treatment.
Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials
- 3,824 trials screened, 41 randomised trials meeting quality criteria included, providing data on 2,672 participants across tendinopathy sites - by far the strongest evidence bearing on injection therapy for an insertional tendinopathy.
- Corticosteroid injection consistently reduced pain in the SHORT term but the effect was REVERSED at intermediate and long term. For lateral epicondylalgia: short-term SMD 1.44 (95% CI 1.17-1.71, p less than 0.0001) favouring injection, but no intervention favoured at intermediate term (SMD −0.40, 95% CI −0.67 to −0.14, p less than 0.003) and long term (−0.31, −0.61 to −0.01, p = 0.05).
- SAFETY: of 991 participants receiving corticosteroid injections in trials that reported adverse events, one (0.1%) sustained a serious adverse event - tendon rupture.
- Short-term efficacy for rotator-cuff tendinopathy was NOT clear, and the authors state explicitly that response to injection should not be generalised because effects vary between tendinopathy sites.
- Limitations for our purposes: NO included trial studied the semimembranosus insertion. The strongest data are from lateral epicondylalgia, so applying these figures to the semimembranosus is an extrapolation the authors themselves caution against - but it is the best guidance available, and there is nothing site-specific to replace it.
Semimembranosus tendonitis after total knee arthroplasty: good outcome after surgery in 6 patients
- Six patients out of more than 700 total knee arthroplasties performed 1992-2000 developed posteromedial knee pain within a year of surgery - roughly 1%, though as a retrospective single-centre series this is an approximate frequency rather than a measured incidence.
- The pain increased on walking and especially on RISING FROM A CHAIR. All six were women, all had good knee motion and well-aligned prostheses, and there were no signs of low-grade infection.
- Injection of local anaesthetic relieved the pain temporarily in all six, confirming the tendon as the source before any surgery.
- Excision of the REFLECTED PORTION of the semimembranosus tendon - which had not been divided during the arthroplasty - relieved pain completely in 5 and improved it in 1.
- Limitations: six patients, no stated follow-up duration, no comparison group and no validated outcome score. It establishes the entity and a plausible remedy, not a success rate.
Proximal hamstring tendinopathy: results of surgical management and histopathologic findings
- IMPORTANT CAVEAT ON SITE: this is PROXIMAL hamstring tendinopathy treated by semimembranosus tenotomy at the ischial origin, not distal insertional disease at the knee. It is included because no comparable surgical series exists for the distal semimembranosus, and because the same authors' warning about site-specific variation applies.
- Case series of 103 operated cases in athletes (58 men, 32 women, 13 bilateral), retrospectively analysed, with an average follow-up of 49 months (range 12-156).
- Results were excellent in 62, good in 30, fair in 5 and poor in 6. Of 90 patients, 80 returned to their pre-symptom level of sport, taking a mean of 5 months (range 2-12 months) - so counsel a months-long, not weeks-long, recovery after surgery.
- Biopsy of 15 operated tendons showed typical morphological features of tendinosis in every specimen, confirming a degenerative rather than inflammatory process - which is the pathological argument against anti-inflammatory-led treatment.
- Limitations: retrospective case series with a non-validated 4-category rating system, no control group, and surgeon-assessed outcomes; the denominator for return to sport (90) differs from the number of cases (103), so the figures are not directly comparable.
Semimembranosus tendinopathy as a recognised cause of posteromedial-corner knee pain and the value of ultrasonography alongside MRI come from the Pobozy review (DOI) - a narrative review reporting no accuracy figures. The injection data - the short-term benefit reversing at intermediate and long term, the effect sizes and confidence intervals, and the 0.1% tendon-rupture rate among 991 injected participants - come from Coombes (DOI), 41 randomised trials and 2,672 participants, none of them at the semimembranosus. The post-arthroplasty presentation and the outcome of excising the reflected arm come from Hendel (DOI), 6 patients. The surgical outcome figures and the tendinosis histology come from Lempainen (DOI), 103 cases of PROXIMAL hamstring tendinopathy - a different site. The five-arm anatomy, the overuse aetiology, the clinical presentation and the posteromedial differential are standard, well-established teaching.
What does not exist: there is no cohort study of distal semimembranosus tendinopathy, and therefore no published prevalence, no recovery time, no success rate for eccentric loading or injection, and no return-to-sport criterion specific to this tendon. No diagnostic-accuracy figure for ultrasound or MRI at this insertion has been reported, so "ultrasound is useful" is expert recommendation rather than measured performance. Any percentage quoted for the outcome of treating this specific tendinopathy would be an extrapolation from other sites, and the Coombes authors caution explicitly against exactly that generalisation.