Introduction

The management of open long bone fractures is well described and has been standardised through a number of well-established guidelines. However, there is no consensus regarding the application of local antibiotics into the open fracture site as a means of reducing infection rates.

The aim of this study was to assess the impact of Covid-19 measures on the rate of surgical site infections (SSI) and subsequent readmissions in orthopaedic patients.

Retrospective, observational study in a level 1 major trauma center comparing rates of SSI in orthopaedic patients who underwent surgery prior to the Covid-19 lockdown versus that of patients who underwent surgery during the lockdown period. A total of 1151 patients were identified using electronic clinical records over two different time periods; 3 months pre Covid-19 lockdown (n=680) and 3 months during the Covid-19 lockdown (n=470). Patients were followed up for 1 year following their initial procedure. Primary outcome was readmission for SSI. Secondary outcomes were treatment received and requirement for further surgeries.

The most commonly performed procedures were arthroplasty and manipulation under anaesthesia with 119 in lockdown vs 101 non-lockdown (p=0.001). The readmission rate was higher in the lockdown group with 61 (13%) vs 44 (6.5%) in the non-lockdown group (p <0.001). However, the majority were due to other surgical complications such as dislocations. Interestingly, the SSI rates were very similar with 24 (5%) in lockdown vs 28 (4%) in non-lockdown (p=0.472). Twenty patients (4.2%) required a secondary procedure for their SSI in the lockdown group vs 24 (3.5%) in non-lockdown (p=0.381). Mortality rate was similar at 44 (9.3%) in lockdown vs 61 (9.0%; p=0.836).

Whilst Covid-19 precautions were associated with higher readmission rates, there was no significant difference in rate of SSI between the two groups.

In recent years, reduction in the length of stay in patients undergoing UKA has gained considerable interest. This has led to development of ‘fast-track' and even day-case protocols aimed at decreasing length of stay (LOS), enhancing post-operative recovery and decreasing post-operative morbidity. One potential barrier to faster discharge and patient recovery is the need for post-operative haemoglobin checks and allogenic blood transfusion; which has been shown to increase LOS. Allogenic blood transfusion itself is not without risk, including immunological reactions, transfusion associated lung injury, infection and transmission of disease, thus reducing blood loss and the need for transfusion is imperative. Currently there is a knowledge gap regarding post-operative transfusion need and blood loss following UKA. We aimed to investigate blood loss and transfusion rates following UKA. Our primary aim was to evaluate the extent of post-operative transfusion need following UKA and identify which patients are at higher risk of needing transfusion.

Following institutional approval, a retrospective analysis of all patients undergoing unicompartmental knee arthroplasty (UKA) at our level one academic university hospital was conducted. Operative records of all patients undergoing primary UKA were reviewed between March 2016 and March 2019. Patients' pre-operative haemoglobin and haematocrit, BMI, co-morbidities, application of tourniquet, tourniquet time, administration of Tranexamic Acid, need for post-operative blood transfusion, hospital length of stay, complications and re-admission were all recorded. Blood loss was estimated using the post-operative haematocrit.

A total number of 155 patients were included. There were 70 females (45%) and 85 males (55%). The mean age was 66±10 years. Median pre-op blood volume was 4700mls (IQR; 4200–5100). Median blood loss was 600 mls (IQR; 400–830). Mean pre-op Haemoglobin was 135±14g/L and mean post-op Haemoglobin was 122±13g/L. No patient had a post-op Haemoglobin under 80g/L (Range 93–154). No patients in our study needed transfusion. A further comparison group of high-blood loss and low-blood loss patients was included in analysis. High-blood loss patients were defined as those losing greater than 20% of their pre-operative blood volume whilst low-blood loss patients were defined as those losing ≤20% of their blood volume. Results of these groups are presented in Table 3. No significance was found between the two groups in patient's demographics and in terms of intra-operative factors including TXA usage (p=0.68) and tourniquet time (p=0.99). There was no difference in terms of post-operative complications (p=1.0), length of stay (p=0.36) or readmission rates (p=0.59).

The results of our study indicated that post-operative haemoglobin and haematocrit check proved unnecessary in all of our patients and could have been omitted from post-operative routines. We conclude that routine post UKA check of haemoglobin and haematocrit can be avoided and be saved for special circumstances depending on patient's physiology.