601 The Effect of the COVID-19 Pandemic on Hip Fracture Mortality in The South West of England

Aim

High perioperative mortality rates in surgical patients suffering concomitant COVID-19 infection have contributed to a fall in elective and urgent surgery. However, data and debate have centred around post hoc infected cases, without the context of the rest of the surgical cohort, and without comparable historical control groups. We aimed to address this by studying patients with a neck of femur fracture. This common condition has a stable incidence with good prospective data collected routinely for the National Hip Fracture Database (NHFD).

Method

We analysed NHFD data for all hospitals within our region over a 3-month period, covering the height of the first outbreak and compared this with the same months in 2019.

Results

The incidence of hip fracture was stable (463 in 2019, 448 in 2020). 30-day mortality was 6.26% in 2019 and 7.14% in 2020 (p = 0.595). In the second cohort, 14 patients tested positive for COVID-19 perioperatively. Of these, 3 died (21%) compared to 29 who tested negative (p < 0.001). Mean time to operation reduced by 1.90 hours, with a significant drop in patients waiting over 36 hours (190 to 85, p < 0.001). There were no significant differences between gender, ASA grade or pre-operative AMTS.

Conclusions

We have confirmed high perioperative mortality for those with COVID-19 infection but have not shown a statistically significant difference in overall mortality from hip fracture during the initial phase of the pandemic. We argue from this data set that the hypothetical risk of surgery during this pandemic may have been overestimated.

Are on-the-day group and save samples required for elective shoulder, hip and knee arthroplasties?

INTRODUCTION

Historically, patients undergoing an elective hip, knee or shoulder arthroplasty regularly required blood transfusions. Improved surgical techniques and perioperative optimisation have significantly decreased the requirement for blood transfusions. Currently, our patients have two group and save samples taken: one six weeks preoperatively and one on admission. This study aims to determine whether a second group and save is required prior to primary elective hip, knee or shoulder arthroplasty.

METHODS

All cases of elective arthroplasty from a single centre were retrospectively analysed over a 16-month period. Each case was reviewed to determine those who had a group and save at preassessment, group and save at the time of the operation and the timing and number of blood products transfused.

RESULTS

A total of 711 elective arthroplasty procedures were completed with 48 patients requiring a transfusion during their admission. 9.9% of hip arthroplasty patients, 3.8% of knee arthroplasty patients and 4.9% of shoulder arthroplasty patients required a transfusion. The majority of the transfusions occurred at least 24 hours postoperatively with 0.84% of patients requiring an intraoperative transfusion.

DISCUSSION

The vast majority of transfusions were delivered more than 24 hours following the procedure, demonstrating that routinely cross-matched blood products are superfluous to requirements. It is our suggestion that a formal group and save be completed only if the need for a blood transfusion is formally established, leading to a significant cost saving, a reduction in clinical work load and patients having to undergo fewer procedures.