Eight trials (n=885) from nine publications were included; one study was prospective and the remaining seven were retrospective. Complete reports on outcomes were available on 609 patients from the included trials.
Only one trial performed analyses adjusted for characteristics of the patients and their injuries. The authors found significant heterogeneity in the outcome measures and the time to follow-up between the studies.
Length of hospital stay (4 studies): only one out of 3 studies found a shorter median length of hospital stay for the primary amputation group (48 days, range: 19 to 116 versus 71 days, range: 28 to 218).
Total rehabilitation time (1 study): the median rehabilitation time after primary amputation was significantly shorter than after the limb salvage procedure (median 12 months, range: 2 to 108 versus 30 months, range: 12 to 72, p<0.009).
Clinical outcomes (5 studies): one study reported that, compared with primary amputation, patients undergoing the limb salvage procedure were more likely to require additional surgery (19.1% versus 5%, p<0.001), undergo rehospitalisation (47.6% versus 33.9%, p=0.002) and develop osteomyelitis (9.4% versus 3.1%, p=0.02). There were no differences in the rates of unhealed soft tissue injury and other infections. One study found that limb salvage patients were more likely to experience leg swelling than those who underwent primary amputation (24 out of 41 versus 2 out of 30, p<0.001). Another study found no differences and 2 studies did not seem to provide a comparison
Failure rate of limb salvage (4 studies): the failure rates for attempted limb salvage resulting in secondary amputation ranged from 9 to 40% in different studies.
Function and quality of life (8 studies): the majority of studies found no significant differences in quality of life between the two groups. Two studies found better functional results for limb salvage patients.
Pain (5 studies): there was no statistically significant difference between the limb salvage and primary amputation groups (RR 1.24, 95% CI: 0.24, 3.12; 4 studies). There was evidence of statistical heterogeneity.
Return to work (7 studies): the pooled meta-analysis did not show any statistically significant differences between the two groups (RR 1.017, 95% CI: 0.96, 1.08).