Aims

Hip fracture commonly affects the frailest patients, of whom many are care-dependent, with a disproportionate risk of contracting COVID-19. We examined the impact of COVID-19 infection on hip fracture mortality in England.

Hip fracture principally affects the frailest in society, many of whom are care dependent, and are disproportionately at risk of contracting COVID-19. We examined the impact of COVID-19 infection on hip fracture mortality in England.

We conducted a cohort study of patients with hip fracture recorded in the National Hip Fracture Database between 1st February 2019 and 31st October 2020, in England. Data were linked to Hospital Episode Statistics to quantify patient characteristics and comorbidities, Office for National Statistics mortality data, and Public Health England's SARS-CoV-2 testing results. Multivariable Cox regression examined determinants of 90-day mortality. Excess mortality attributable to COVID-19 was quantified using Quasi-Poisson models.

Analysis of 102,900 hip fractures (42,630 occurring during the pandemic) revealed that amongst those with COVID-19 infection at presentation (n=1,120) there was a doubling of 90-day mortality; hazard ratio (HR) 2.05 (95%CI 1.86–2.26), while for infections arising between 8–30 days after presentation (n=1,644) the figure was even higher at 2.52 (2.32–2.73). Malnutrition [1.44 (1.19–1.75)] and non-operative treatment [2.89 (2.16–3.86)] were the only modifiable risk factors for death in COVID-19 positive patients. Patients with previous COVID-19 initially had better survival compared to those who contracted COVID-19 around the time of their hip fracture; however, survival rapidly declined and by 365 days the combination of hip fracture and COVID-19 infection was associated with a 50% mortality rate. Between 1st January and 30th June 2020, 1,273 (99.7%CI 1,077–1,465) excess deaths occurred within 90 days of hip fracture, representing an excess mortality of 23% (20%–26%), with most deaths occurring within 30 days.

COVID-19 infection more than doubled early hip fracture mortality; the first 30-days after injury were most critical, suggesting that targeted interventions in this period may have most benefit in improving survival.

Aims: Data collected on knee replacements from 93 hospitals in England were analysed to identify risk factors for surgical site infection. Methods: Prospective surveillance data collected from hospitals in England to calculate surgical site infection rates and potential risk factors Results: There were 242 (1.6%) infections in 15427 primary TKR, with a 0.11% joint infection rate, and 36 (3.6%) infections in 988 revisions, with a signiþcantly higher joint infection rate of 0.35%. There was considerable variation between hospitals. Multivariable logistic regression found signiþcant association between risk of SSI and type of procedure (p< 0.01), the hospital where the procedure was performed, male sex (p< 0.001), age (p< 0.001) and wound class (p< 0.05). The mean length of stay in primary TKRñs increased from 9 days to 18 days with an infection; and in revision procedures from 12 to 24 days. The median time to diagnosis for SSI was 7.5 days. Staphylococcus aureus accounted for 33%, 66% of which were methicillin resistant (MRSA). Conclusions: There is signiþcant inter-hospital variation in the incidence of SSI following knee replacements in England. Signiþcant risk factors include revision procedures. Male sex, age and wound class. SSIñs doubled the length of post-operative stay and MRSA accounted for 21% of organisms.

Aims: To calculate SSI rates and potential risk factors for primary total hip replacements (THR), hip hemiarthroplasties (HH), revision THR’s, and revision HH’s in England. Methods: Demographic, operative, and infection data was collected prospectively from 104 hospitals in England over 4 years to calculate SSI rates and potential risk factors. Results: There were 353 (2.3%) infections in 15697 THR’s with a 0.2% joint infection rate; 248 (4.6%) infections in 5456 HH procedures (0.9% joint). For revision THRs there were 92 (3.6%) infections in 2563 procedures, and 11 (5.6%) infections in 197 revision HH’s. Staphylococcus Aureus was identified in 52%; 58% were MRSA. With multivariable analysis, only inter-hospital variation (p< 0.001) and ASA score (p< 0.001) remained significant. Mean time to detection of infection varied from 9.2 days (superficial) to 11.3 days (joint). Mean length of stay increased from 9 to 14 days in THR’s; from 14 to 30 days for revision THR’s; from 19 to 34 days for HH’s and from 20 to 28 days for revision HH’s with an infection.

Conclusions: Infection rates are within acceptable levels, however there is a significant inter-hospital variation in infections following hip prosthesis in England. ASA score remains a significant risk factor and MRSA accounted for 30% of organisms.

Data collected on total knee replacements (TKR) from 77 hospitals in England were analysed to identify risk factors for surgical site infection (SSI).

Demographic, operative, and infection data were collected prospectively over a four-year period by the Nosocomial Infection National Surveillance Scheme.

There were 213 (1.8%) infections reported in 11552 primary TKR of which 82% were superficial, 10% deep incisional, and 8% joint/bone infections. The incidence of SSI in 687 revision of TKRs was 4.1% (71% superficial incisional, 18% deep incisional and 11% joint/bone). In the single variable analysis of primary TKRs, significant risk factors were male sex (p< 0.01), age (p< 0.001), ASA score (p< 0.001), wound class (p< 0.001) and NNIS risk index (p< 0.001). In revision of TKRs, only age (p< 0.01) and pre-operative hospital stay of more than one day (p< 0.02) were found to be significant. Significant risk factors with multi-variable logistic regression were type of procedure (TKR or revision TKR), hospital where the procedure was performed, male sex, and age. The mean length of stay in primary TKRs was 10 days (19 days with SSI) and 12 days in revision TKR (22 days with SSI). The median time to diagnosis for superficial SSI was 7 days for superficial SSIs, 9 days for deep incisional SSIs and 7.5 days for joint/bone infections. Staphylococcus aureus accounted for 35% of the infections and nearly one third of these were methicillin resistant (MRSA).

There is significant inter-hospital variation in the incidence of SSI following total knee replacement. Revision TKR procedures are associated with a significantly higher incidence of SSI than primary TKRs (p< 0.001). Male sex and age are also important risk factors. Patients with SSI had a length of post-operative stay approximately twice that of those without SSI.