Table 5. Summary of results of all 21 randomized controlled trials (RCTs) included to date.
| RCT | Primary endpoint | Secondary endpoints | |||
| Complete wound closure | Time to complete wound closure | Reduction in wound size*1 | Mortality | Other adverse events*2, *3 | |
| Armstrong 2005 (e2) | +++ | +++ | n/a | (+) | (+) |
| Bee 2008 (12) | (+) | n/a | n/a | (–) | (–) |
| Blume 2008 (8) | +++ | +++ | +++ | (+) | (+) |
| Braakenburg 2006 (e9) | n/a | (+) | (+) | (+) | (–) |
| Chio 2010 (9) | n/a | n/a | n/a | n/a | (+) |
| Eginton 2003 (e3) | n/a | n/a | +++ | n/a | n/a |
| Ford 2002 (e4) | (–) | n/a | (+) | n/a | (–) |
| Joseph 2000 (e5) | n/a | +++ | +++ | n/a | +++ |
| Keskin 2008 (10)*4 | n/a | n/a | n/a | n/a | n/a |
| Llanos 2006 (e10) | n/a | n/a | +++ | n/a | n/a |
| Mody 2008 (13) | (–) | (–) | n/a | n/a | (–) |
| Moisidis 2004 (e6) | n/a | n/a | n/a | n/a | n/a |
| Mouës 2004 (e7) | n/a | (+) | +++ | n/a | n/a |
| Perez 2010 (14) | (–) | +++ | n/a | n/a | –- |
| Saaiq 2010 (15) | +++ | +++ | n/a | 0 | n/a |
| Sepúlveda 2009 (16) | n/a | +++ | n/a | 0 | n/a |
| Stannard 2006a (e11)*5 | n/a | n/a | n/a | n/a | (–) |
| Stannard 2006b (e11) | n/a | n/a | n/a | n/a | 0 |
| Stannard 2009 (11)*6 | n/a | n/a | n/a | n/a | –- |
| Vuerstaek 2006 (e12) | n/a | +++ | +++ | (–) | (–) |
| Wanner 2003 (e8) | n/a | (+) | (+) | n/a | n/a |
+++: statistically significant difference in favor of negative-pressure wound therapy (NPWT); –-: statistically significant difference in favor of comparator treatment; (+): insignificant difference in favor of NPWT; (–): insignificant difference in favor of comparator treatment; 0: no difference; n/a: not available
*1Wound size: Surface area or volume, statistically significant results reported primarily
*2Pain: Braakenburg 2006 referred to a study on pain but did not report its results; Vuerstaek 2006 reported a statistically significant benefit of NPWT, but there were already statistically significant differences in pain between the two groups initially (pain was lower at baseline in the NPWT group). This was not suitably taken into account in the evaluation.
*3Further results that were only reported singly and were therefore not included in the table: amputations: Armstrong 2005: (+); Blume 2008: +++; Braakenburg 2006 (+). Quality of life: Keskin 2008: –-; Stannard 2009: +++; Vuerstaek 2006: –- (in the first week of treatment)
*4Keskin 2008: fear during the first 10 days
*5Stannard 2006a: trial on hematomas; Stannard 2006b: trial on incisions in fractures
*6Stannard 2009: statistically significant difference after 3, 6 and 9 months