Howell 2011.

Methods	Study design: randomised controlled trial Ethics and informed consent: not reported Sample size calculation: yes Follow‐up period: 12 months ITT analysis: all participants completed the study. Funding: the study was supported by KCI, the manufacturer of the negative pressure device.
Participants	Location: NYU Hospital for Joint Disorders, New York, NY, USA Intervention group: n = 24control group: n = 36 Mean age: not reported Inclusion criteria: patients undergoing unilateral or bilateral primary total knee arthroplasty who were obese (BMI > 30), who met criteria of increased risk for postoperative wound drainage and who were prescribed enoxaparin sodium for deep vein thrombosis prophylaxis Exclusion criteria: patient refusal to participate in the study, revision total knee replacement, prior knee surgery (except arthroscopy), and patients with documented diabetes mellitus
Interventions	Aim/s: to compare the number of days to dry wound in a negative pressure dressings group compared with a static pressure dressings group Intervention/s in both groups: "all patients received three doses of peri‐operative intravenous antibiotics and were maintained on subcutaneous DVT prophylaxis for 30 days after surgery" Group 1 (NPWT) intervention: "subsequent to the closure of the surgical incision, a negative pressure dressing (VAC Therapy, Kinetic Concepts Inc., San Antonio, Texas) was applied under sterile conditions. A medical grade open cell polyurethane ether foam (pore size of 400‐600 micrometers) was cut into the shape of a rectangle approximately 5 cm in width and a length sufficient to cover the entire linear wound. The knee was held in 151° of flexion, and the foam was secured over the incision by the application of a specialized adhesive drape, provided in the NPWT system. An evacuation tube with side ports was embedded within the reticulated foam, allowing negative pressure to be applied equally over the entire wound bed. The foam‐evacuation tube complex attached to a programmable vacuum pump applied a −125 mmHg continuous vacuum pressure to the wound. The NPWT dressing remained in place for a 48‐hour period, after which time clean, dry gauze dressings were applied and changed on daily basis until the wound was dry" Group 2 (SPD) intervention: "patients in the control arm had their surgical wound covered in the operating room with a sterile, dry gauze dressing that was held in place with a perforated, stretchable cloth tape. This initial dressing remained in place for 48 hours after which time clean, dry gauze dressings were applied and changed on a daily basis until the wound was dry" Study date/s: not stated
Outcomes	days to dry wound deep wound infection blister formation Time points: participants followed up for 12 months postsurgery.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Comment: not described
Allocation concealment (selection bias)	Low risk	Quote: "randomised with blinded envelopes to either the treatment with negative pressure wound therapy group or a control group using sterile gauze"
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Evidence for participants: not possible Comment: unlikely to affect outcomes Evidence for personnel: not possible Comment: unlikely to affect outcomes Comment: difference in appearance of dressings made blinding impossible.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Evidence: not described
Incomplete outcome data (attrition bias) All outcomes	Low risk	Evidence: 51 participants were randomised, and 51 completed the study.
Selective reporting (reporting bias)	Low risk	Comment: the prespecified clinical outcomes were presented in Table 3 in the trial report, and a post hoc analysis of blister occurrence was shown in Table 4. Infection rates were reported in the results section of the trial report. We could not find a published protocol.
Other bias	High risk	No baseline data were presented. In addition, groups contained unequal numbers, which could indicate undisclosed losses in 1 group.