Methods

Study design: randomised controlled trial, two groups

Study grouping: parallel group

Setting: participants were recruited from physicians and physical therapists referring participants to the Health Department at the Olympic Training Center in Oslo, Norway

Timing: March 2001 to September 2004
Intervention: open surgical excision combined with eccentric exercises versus physical therapy with eccentric exercises alone

Sample size: 15 participants required per group to have 90% power to detect a difference of mean difference of 13 points on the Victorian Institute of Sport Assessment (VISA) score between groups in the primary endpoint; overall type‐1 error rate was set at 5%

Analysis: intention‐to‐treat analysis planned and executed

Participants

Number of participants

• 40 randomised (20 to eccentric training, 20 to open surgical excision)

• At six‐month follow‐up there were 18/20 in the open surgical excision group and 20/20 in the eccentric exercise group included in the analysis.

• At 12‐month follow‐up there were data for 20/20 (100%) for the open surgical excision group and 20/20 (100%) for eccentric exercise group.

Inclusion criteria

• History of exercise‐related pain at the proximal part of the patellar tendon or the patellar insertion for at least three months, with Blazina scale (Lian modified) 3B symptoms ‐ meaning the patient had to have pain during and after activity and had to be unable to participate in sports at the same level as before the onset of pain

• Tenderness to palpation corresponding to the painful area

• Magnetic Resonance Imaging (MRI) ‐ thickening and increased signal intensity changes corresponding to the proximal part of the patellar tendon or the patellar insertion

Exclusion criteria

• History of knee or patellar tendon surgery

• Had an inflammatory or degenerative joint condition

• Less than 18 years old

• Unable to understand oral and written Norwegian

• Unwilling to undergo surgery (five participants)

Baseline characteristics: there was no demographic difference between groups. Listed recorded demographics include age, gender, height, weight, baseline VISA score, duration of symptoms, length of participation in sport and the amount or type of training.

Open surgical excision group

• Mean age: 30 (SD 8) years (3 female, 17 male)

• Symptom duration: 35 (SD 30) months

• Mean training hours: 12.2 (SD 7.3)

• Mean pain score: 4.3 (SD 2.3)

• Mean VISA functional score: 31 (SD 15)

Eccentric exercises group

• Mean age: 31 (SD 8) years (2 female, 18 male)

• Symptom duration: 33 (SD 28) months

• Mean training hours: 11.2 (SD 7.2)

• Mean pain score: 3.9 (SD 2.7)

• Mean VISA functional score: 29 (SD 16)

Interventions

Open surgical excision

Surgery involved a 5 cm longitudinal midline incision from the inferior pole of the patella distally. A tourniquet was not used. The paratenon was split longitudinally and any pathologic paratenon tissue was removed. The tendon was split longitudinally in the midline to expose the deepest layers, and tissue that appeared abnormal was removed, or if not seen macroscopically the area calculated from the MRI was removed. Once the sutures were removed the participants were exposed to the same physiotherapist and eccentric exercises as the non‐operative group. Two surgeons performed the operations.

Prior to starting the eccentric exercises participants gradually increased activities, focusing on knee range‐of‐motion and walking without crutches.

Eccentric exercise group

Involved use of a 25‐degree decline board at home, where the downward (eccentric) component was performed on the affected leg and the upward (concentric) component was performed on the asymptomatic leg. The participant was instructed to perform the squat to 90 degrees with the back in a vertical position. The patient was instructed to take two seconds for the eccentric component. It involved twice‐daily sessions of three sets of 15 repetitions for a total duration of 12 weeks, with weekly supervision by a physiotherapist. No warm‐up was performed. Six weeks after surgery, participants were exposed to the same rehabilitation/eccentric exercises programme as the eccentric exercise group, except they were not allowed to tolerate pain.

Both groups were allowed pain relief freely.

Outcomes

Outcomes were assessed at baseline, 12 weeks, 6 months and 12 months by the trial authors. We report 6‐ and 12‐month outcomes in this review.

Outcomes included in the trial

• Function: assessed by VISA zero‐to‐100 scale; 100 is full, pain‐free function

• Participant global assessment of treatment: measured on an 11‐point numerical scale ranging from ‐5 (maximal deterioration in symptoms) to +5 (maximal improvement in symptoms)

• Participant‐reported overall treatment satisfaction, measured using four categories: no symptoms, improved, no change, or worse

• Return to sport: measured using a four‐point categorical scale, where 0 is no sport, 1 (reduced level), 2 (full training but some symptoms), 3 (full training and no symptoms).

• Pain with jumping: assessed on 0‐ (no pain) to 10‐point (maximum pain) visual analogue scale (VAS) for standing jumps and counter‐movement jumps

• Height of standing jumps and counter‐movement jumps, in cm

• Strength: maximum load during leg extension using a leg press, in kg

Outcomes included in this review

• Pain: mean pain with standing jump on 0‐ to 10‐point VAS

• Function: mean score on zero‐to‐100 VISA scale

• Participant global assessment: reported as proportion who reached full training and no symptoms

• Proportion with adverse events

Identification

Sponsorship source: one or more of the authors received grants or outside funding from Norwegian Eastern Health Corporate, Royal Norwegian Ministry of Culture, Norwegian Olympic Committee and Confederation of Sport, Norsk Tipping AS, and Pfizer AS.

Author's name: Roald Bahr

Institution: Norwegian School of Sport Sciences

Email: [email protected]

Address: Oslo Sports Trauma Research Center, Department of Sports Medicine, Norwegian School of Sport Sciences, P.O. Box 4014 Ullevaal Stadion, 0806 Oslo, Norway.

Notes

Trial registration: none found

Data analysis: mean (95% confidence interval (CI)) function and participant global assessment of success at 6 and 12 months were extracted from the graph using PlotDigitizer (automeris.io/WebPlotDigitizer/).

Adverse events: one participant in the surgery group developed chronic quadriceps pain. Tendon rupture did not appear to be measured.

Withdrawals: none

Cross‐overs: 5/20 in the eccentric exercise group had surgery.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

"A randomisation sequence to surgical treatment or eccentric training (in blocks of four) was created by our statistician prior to the start of the study". It is likely this was adequate.

Allocation concealment (selection bias)

Low risk

"Sealed opaque envelopes used", and opened after a participant was enrolled in the study. This was probably sufficient to conceal treatment allocation.

Blinding of participants and personnel (performance bias)
All outcomes

High risk

Neither the participants or the investigators were blinded with regard to the group allocation. There is potential for performance bias.

Blinding of outcome assessment (detection bias)
Self reported outcomes

High risk

Given that participants were allocated to surgery or exercises, blinding would be difficult, thus there is a potential for detection bias in reporting pain, function, and global evaluation.

Blinding of outcome assessment (detection bias)
Assessor reported outcomes (tendon rupture)

Low risk

This outcome was not measured in the study.

Incomplete outcome data (attrition bias)
All outcomes

Low risk

5/20 participants from the eccentric exercises group crossed over to open surgical excision, but final score before surgery was carried forward to 12 months in the eccentric exercise group, which had the potential to attenuate any benefit of surgery if one existed. Post hoc analysis in the trial suggests no change in outcome over time with these participants.

Selective reporting (reporting bias)

High risk

Multiple outcomes reported. VISA score is the primary outcome and reported as such. Pain was only reported at the 12‐month time point. There is no protocol publication and it is unclear if adverse events, including tendon rupture, were measured.

Other bias

Low risk

None apparent

Methods

Study design: randomised controlled trial, two groups

Study grouping: parallel group

Setting: participants were recruited from Capio Artro Clinic in Stockholm, Sweden

Timing: not reported
Intervention: sclerosing injection (polidocanol) versus arthroscopic surgery. Post intervention, both groups were allowed to weight bear as tolerated and increase loading as symptoms allowed from two weeks.

Sample size: 15 participants required per group to have 90% power to detect a difference of mean difference of 13 points on the VISA score between groups in the primary endpoint; overall type‐1 error rate was set at 5%

Analysis: intention‐to‐treat analysis

Participants

Number of participants

• 52 participants were randomised (26 to sclerosing injection (Polidocanol) and 26 to arthroscopic surgery).

• Data were available for 25 (96%) for the injection group and 25 (96%) for the arthroscopic surgery group at 12‐month follow‐up.

Inclusion criteria

• History and clinical examination consistent with proximal patella tendinopathy, long duration of symptoms, an athlete

• Imaging‐ultrasound showing the pathological changes in the thickened proximal patellar tendon were recorded

Exclusion criteria: acute presentation of pain

Baseline characteristics: there were no demographic differences.

Arthroscopic surgery group

• Mean age: 26.6 (SD 7.6) years (2 female, 24 male)

• Symptom duration: 24 (SD 15.5) months

• Mean training hours: not reported

• Mean pain score: 76.5 (SD 13.6)

• Mean VISA functional score: not reported

Sclerosing injection (Polidocanol) group

• Mean age: 27 (SD 7.6) years (2 female, 24 male)

• Symptom duration: 20 (SD 10.4) months

• Mean training hours: not reported

• Mean pain score: 69 (SD 17.3)

• Mean VISA functional score: not reported

Interventions

Both interventions were performed under ultrasound and were doppler‐guided.

Arthroscopic surgery

Arthroscopy was performed under local anaesthetic, using anterolateral and anteromedial portals. No tourniquet was used. The patella tendon insertion into the patella was identified following a routine arthroscopic examination of the knee. A 4.5 mm shaver was utilised to destroy only the region with high blood flow and nerves adjacent to the tendinosis changes on the dorsal side of the tendon. No tendon tissue was resected and the Hoffa fat pad was saved. Portals were closed with tape. No information was given on who performed the surgeries.

Sclerosing injection (Polidocanol)

A 0.7 mm x 50 mm needle, connected to a 2 mL syringe, was utilised. Volumes of 0.1 mL to 0.2 mL were injected into the regions with local neovascularisation/high blood flow dorsal to the proximal patellar tendon. A maximum of three injections were given at six‐week intervals. All injections were performed by the same, single sonographer.

Cointerventions

Both groups were allowed to weightbear fully post‐treatment, and there was no specific rehabilitation protocol or preclusion of activities.

Outcomes

Outcomes were assessed at baseline, 2 weeks, 3 weeks, 6 months and 12 months by the trial authors, but reported at only 12 months. We report outcomes at 12 months in this review.

Outcomes included in the trial

Pain ‐ rest

• Pain at rest: mean pain assessed on a VAS of 0 mm (no pain) to 100 mm (maximal pain)

• Pain during activity: mean pain assessed on a VAS of 0 mm (no pain) to 100 mm (maximal pain) with sporting activity

• Participant‐reported satisfaction with outcome: mean satisfaction assessed on a VAS of 0 mm to 100 mm (maximal satisfaction)

Outcomes included in this review

• Pain during activity: mean pain assessed on a VAS of 0 mm to 100 mm

• Participant‐reported satisfaction, mean satisfaction assessed on a VAS of 0 mm to 100 mm

Identification

Sponsorship source: nil reported

Country: Sweden

Setting: Capio Artro Clinic, Stockholm

Author's name: Dr Lotta Willberg

Institution: Capio Artro Clinic, Stockholm

Email: [email protected]

Address: CapioArtro Clinic, StockholmSports Trauma Research Centre, Karolinska Institutet, Sophiahemmet, Valhallavägen91, S‐114 86 Stockholm, Sweden

Notes

Trial registration: none found.

Data analysis: mean (SD) pain and participant global assessment of success at 12 months was extracted from the study table.

Adverse events: no adverse events were reported.

Withdrawals: 1/26 participants withdrew in both groups.

Cross‐overs: none

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

The method of sequence generation was not provided.

Allocation concealment (selection bias)

Unclear risk

Concealed envelopes were used, but participants selected their own envelope.

Blinding of participants and personnel (performance bias)
All outcomes

High risk

Study participants and investigators were not blinded and were aware of the treatment; therefore there is a risk of performance bias.

Blinding of outcome assessment (detection bias)
Self reported outcomes

High risk

The participants were not blinded with regard to the group allocation. There is potential for detection bias for self‐reported outcomes (pain, function, global evaluation) and performance‐related outcomes (e.g. standing jump) due to knowledge of the allocated interventions by participants. The assistant collecting outcome scores was blinded to group allocation, however the data was recorded by participants who were not blinded, therefore there is a risk of detection bias.

Blinding of outcome assessment (detection bias)
Assessor reported outcomes (tendon rupture)

Low risk

Not relevant as none measured

Incomplete outcome data (attrition bias)
All outcomes

Low risk

1/26 participants from both groups withdrew.

Selective reporting (reporting bias)

High risk

Complete data set not available for all mentioned time points. The study did not report known outcome scores for patella tendinopathy (i.e. VISA) but may have recorded these.

Other bias

Low risk

None apparent