Study characteristics and risk of bias assessment
| Comparator group: | (1) Non-surgical | (2) Pharmacological | (3) Surgical | |||||||
| Herrlin et al 51 66 | Katz et al 54 72 73 | Østerås et al 74 75 | Yim et al 53 | Gauffin et al 42 76 | Kise et al 43 | van de Graaf et al 45 | Vermesan et al 55 | Sihvonen et al 36 62 71 77 78 | Roos et al 44 | |
| Comparator (control) | Physiotherapy | Physiotherapy | Physiotherapy | Physiotherapy | Physiotherapy | Physiotherapy | Physiotherapy | Steroid injection | Placebo APM | Sham APM |
| Sample (randomised) | 99 | 351 | 17 | 108 | 150* | 140* | 321* | 114 | 146 | 44* |
| Sample (analysed) | 90 | 320 | 17 | 102 | 130* | 129* | 289* | 98 | 146 | 42* |
| Lost to follow-up, n (%) | 9 (9.1) | 31 (8.8) | 0 (0) | 6 (5.6) | 20 (13.3) | 11 (7.9) | 13 (4.0) | 16 (14.0) | 0 (0) | 2 (4.5) |
| Number of centres | Single | Multiple | Multiple | Single | Single | Multiple | Multiple | Single | Multiple | Multiple |
| Age mean (SD) (controls) | 55 (5.5) | 58 (6.8) | 47 (10.4) | 58 (11) | 54 (6) | 50 (6.2) | 57.3 (6.8) | 58 (7.8) | 52 (7) | 46 (5.7) |
| Female, % (controls) | 37 | 57 | 11 | 77 | 25 | 39 | 50 | 46 | 38 | 61 |
| Minimum symptom duration | 2 months | 4 weeks | 3 months | 6 weeks | 3 months | 2 months | NR | NR | 3 months | 2 months |
| Osteoarthritis grade | Ahlbäck 0–1 | K-L 0–3 | K-L 0–2 | K-L 0–1 | K-L 0–2* | K-L 0–3* | K-L 0–2* | NR | K-L 0–1 | K-L 0–2* |
| Intervention | APM | APM | APM | APM | APM | APM | APM | APM | APM | APM |
| Outcome measures (PROMs) | KOOS. Lysholm. VAS Pain. Tegner. | WOMAC (physical). SF-36 (physical). KOOS Pain. | KOOS VAS Pain | Lysholm. VAS Pain. | KOOS. EQ-5D. | KOOS-4. SF-36 (physical). SF-36 (mental). | IKDC. VAS Pain. SF-36 (physical). EQ-5D (NR). Tegner. | OKS. | Lysholm. WOMET. 15D. VAS Pain. | KOOS. EQ-5D. SF-36 (physical). SF-36 (mental). |
| Endpoints (months) | 6, 24, 60 | 6, 12 | 3 | 3, 6, 12, 24 | 12, 36 | 3, 12, 24 | 3, 6, 12, 24 | 1, 12 | 6, 12, 24 | 3, 24 |
| Cross-over †, % | 28.2 | 30.2 | NR | 1.9 | 21.3 | 18.6 | 29.0 | 20.8 | 6.6 | 36.0 |
| Risk of bias (Cochrane Collaboration risk-of-bias tool)38 | ||||||||||
| Random sequence generation (selection bias) | Unclear | Low | Unclear | Unclear | Unclear | Low | Low | Unclear | Low | Low |
| Allocation concealment (selection bias) | Unclear | Low | Unclear | Unclear | Low | Low | Unclear | Unclear | Low | Low |
| Blinding of participants and researchers (performance bias) | High | High | High | High | High | High | High | High | Low | Low |
| Blinding of outcome assessment (detection bias) | High | High | High | High | High | High | High | High | Low | Low |
| Incomplete outcome data (attrition bias) | Unclear | Low | Low | Low | High | Low | Low | High | Low | Low |
| Selective reporting (reporting bias) | Unclear | High | Unclear | Unclear | Unclear | High | Unclear | Unclear | Low | Low |
| Other bias (cross-over,† baseline imbalance) | High | High | Unclear | Low | High | High | High | High | Low | High |
Risk of bias assessment (Cochrane Collaboration risk-of-bias tool).38
Further details including a full description of the intervention and comparator treatments, funding and location may be found in online supplementary appendix 4.
*Previously unpublished subgroup data (K-L 0–1) received.
†Underwent APM intervention after randomisation to control treatment.
APM, arthroscopic partial meniscectomy; EQ-5D, EuroQol five dimensions questionnaire; IKDC, International Knee Documentation Committee; K-L, Kellgren Lawrence grade; KOOS, Knee injury and Osteoarthritis Outcome Score; NR, not reported; OKS, Oxford Knee Score; PROMs, patient-reported outcome measures; SF-36, 36 -Item Short Form Health Survey; VAS, visual analogue pain score; WOMAC, Western Ontario and McMaster Universities Osteoarthritis Index; WOMET, Western Ontario Meniscal Evaluation Tool; .