A recent systematic review and narrative synthesis has found several tools that demonstrate “acceptable-to-outstanding discrimination” for predicting key outcomes following major lower limb amputation. However, authors Ryan A Preece (Royal Gwent Hospital, Cardiff, UK) and colleagues note that “additional robust external validation” is needed to strengthen the clinical applicability of the prediction tools assessed. This research appeared online in the European Journal of Vascular and Endovascular Surgery (EJVES).
Considering the rationale behind their study, Preece et al stress that “the decision to undertake a major lower limb amputation can be complex”. They remark that, in the literature, previous studies have highlighted that understanding of factors that lead to adverse outcomes following amputation is “lacking,” adding that poor decision making around amputation can “dramatically reduce quality of life, and also prove to be very costly to healthcare providers”. Therefore, the research team decided to identify and evaluate the performance of risk prediction tools in estimating mortality, morbidity, and other outcomes following amputation.
Preece and colleagues performed a systematic review, searching the MEDLINE, Embase, and Cochrane databases to identify studies reporting on risk prediction tools that predict outcomes following amputation. They write: “Outcome measures included the accuracy of the risk tool in predicting a range of postoperative complications, including mortality (both short- and long-term), perioperative morbidity, need for reamputation, and ambulation success.” Following the review, they performed a narrative synthesis of the collected studies. Reporting in EJVES, they specify that nine tools predicted mortality outcomes, with two predicting morbidity, two for successful ambulation outcomes, and one for the necessity for revision surgery.
The investigators’ search identified 518 database records, including 12 observational studies, evaluating 13 risk prediction tools in a total cohort of 61,099 amputations. Most mortality prediction tools demonstrated “acceptable discrimination performance,” the authors relay, noting C statistic values ranging from 0.65 to 0.81. They add that tools estimating the risk of postoperative complications and necessity for reamputation “also performed acceptably,” with C values of 0.65–0.74 and 0.72, respectively. Finally, Preece et al communicate that the Blatchford Allman Russell tool demonstrated “outstanding discrimination” for predicting functional mobility outcomes postamputation (0.94). Overall, however, they highlight that “most studies were at high risk of bias with poor external validity”.
In their conclusion, the authors stress that, while the tools assessed show “promising performance,” there is a “significant lack” of external validation studies, which “urgently needs addressing with prospective studies”. Based on this gap in the available data, they urge further study before implementation in clinical practice and also that “further efforts need to be invested into delivering the calculations through convenient, user-friendly platforms”.