Clinical Effects of Tourniquet Use for Nontraumatic Major Lower Limb Amputation: A Two-Center Retrospective Cohort Study

Owen Richards; Yusuf Cheema; Brenig Gwilym; Graeme K Ambler; Christopher P Twine; David C Bosanquet

doi:10.1016/j.avsg.2023.07.096

Clinical Effects of Tourniquet Use for Nontraumatic Major Lower Limb Amputation: A Two-Center Retrospective Cohort Study

Ann Vasc Surg. 2023 Jul 14:104:53-62. doi: 10.1016/j.avsg.2023.07.096. Online ahead of print.

Authors

Owen Richards¹, Yusuf Cheema², Brenig Gwilym³, Graeme K Ambler⁴, Christopher P Twine⁴, David C Bosanquet³

Affiliations

¹ School of Medicine, Cardiff University, Cardiff, UK; South East Wales Vascular Network, Aneurin Bevan University Health Board, Royal Gwent Hospital, Newport, UK. Electronic address: Owen.richards4@wales.nhs.uk.
² School of Medicine, Cardiff University, Cardiff, UK.
³ South East Wales Vascular Network, Aneurin Bevan University Health Board, Royal Gwent Hospital, Newport, UK.
⁴ Department of Vascular Surgery, North Bristol NHS Trust, Bristol, UK; Centre for Surgical Research, University of Bristol, Bristol, UK.

PMID: 37453468
DOI: 10.1016/j.avsg.2023.07.096

Abstract

Background: To investigate the effect of tourniquet use on outcomes after major lower limb amputation (MLLA) due to peripheral arterial disease or complications from diabetes mellitus.

Methods: In this 2-center retrospective observational study, vascular patients who underwent MLLA between January 1, 2016 and December 31, 2020 at 2 UK hospitals were identified using operating theater databases. Hospital databases were used to access medical records, operation notes, and laboratory reports. The use of a tourniquet in each MLLA was noted. The primary outcome was postoperative hemoglobin (Hb) drop (g/L). Secondary outcomes were units of allogeneic blood transfused perioperatively, 90-day revision rates, 90-day wound breakdown rates, surgical site infection (SSI) rates (at 30 days), and 90-day mortality. A follow-up index (a measure of follow-up completeness) was calculated for all 30-day and 90-day outcomes.

Results: Four hundred seventy two patients underwent MLLA, of which 124 had a tourniquet applied. The median postoperative Hb drop was significantly lower in the tourniquet group compared to the nontourniquet group (13 [interquartile range 5-22] g/L vs. 20 [interquartile range 11-28] g/L; P ≤ 0.001). Thirty three point one percent (41) of tourniquet patients received a blood transfusion perioperatively, compared to 35.6% (124) of nontourniquet patients (P = 0.82). Sixteen percent (76) of patients required surgical revision within 90 days, with no significant difference between the tourniquet and nontourniquet group (20.2% tourniquet vs. 14.7% no tourniquet; P = 0.15). SSI rates (12.0% tourniquet vs. 10.6% no tourniquet, P = 0.66) and 90-day mortality (6.5% tourniquet vs. 10.1% no tourniquet; P = 0.23) were similar. Multivariable regression demonstrated that tourniquet use was independently associated with a reduced hemoglobin drop (β = -4.671, 95% confidence interval -7.51 to -1.83, P ≤ 0.001) but was not associated with wound breakdown, revision surgery, or SSI. Hypertension, SSI, and below-knee amputation using the skew flap technique were all significant predictors of revision surgery. All follow-up indices were ≥ 0.97.

Conclusions: Tourniquet use in MLLA was associated with a significantly lower fall in postoperative Hb without evidence of harm in terms of SSI, wound breakdown/revision rates, or mortality.