Establishing a COVID-secure site for elective surgery during the COVID pandemic: An observational study

Abstract

Introduction

Maintaining timely and safe delivery of major elective surgery during the COVID-19 pandemic is essential to manage cancer and time-critical surgical conditions. Our NHS Trust established a COVID-secure elective site with a level 2 Post Anaesthetic Care Unit (PACU) facility. Patients requiring level 3 Intensive Care Unit admission were transferred to a non-COVID-secure site. We investigated the relationship between perioperative anaesthetic care and outcomes.

Materials and methods

All consecutive patients undergoing major surgery at the COVID-secure site between June and November 2020 were included. Patient demographics, operative interventions and 30-day outcomes were recorded. Multivariate logistic regression was used to determine the odds ratio of outcomes according to PACU length of stay and the use of spinal or epidural anaesthesia, with age, sex, malignancy status and American Society of Anesthesiologists grade as independent co-variables.

Results

There were 280 patients. PACU length of stay >23h was associated with increased 30-day complications. Epidural anaesthesia was associated with PACU length of stay >23h, increased total length of stay, increase hospital transfer and 30-day complications. Two patients acquired nosocomial COVID-19 following hospital transfer.

Discussion

Establishing a separate COVID-secure site has facilitated delivery of major elective surgery during the COVID-19 pandemic. Choice of perioperative anaesthesia and utilisation of PACU appear likely to affect the risk of adverse outcomes.

Provenance and Peer review: Unsolicited contribution; Peer reviewed; Accepted for publication 20 June 2021.

Introduction

As a result of the COVID-19 pandemic and at the time of writing, nearly 10 million people are awaiting a surgical procedure in the UK, and close to 140,000 in England have been waiting for over a year for their operation (The Lancet Rheumatology 2021). Elective waiting times are increasing in duration, with some hospitals rejecting new referrals (GP Online 2020, Griffin 2020). Those awaiting time-critical surgery are likely to be considered ‘high-risk’ and advised to self-isolate under government guidelines.

There are serious implications for patients undergoing major surgery while co-infected with SARS-CoV-2 in the perioperative period (Nepogodiev et al 2020). Maintaining timely and safe delivery of major urgent elective surgery throughout the pandemic is essential if cancer and other time-critical surgical conditions are to be managed optimally. In response, some centres have designated ‘cold’ or ‘green’ hospital pathways in an attempt to continue elective surgery in a ‘COVID-secure’ environment. It is not yet known whether such pathways are effective or what types of perioperative care are optimal since there is a sparsity of data from such a context.

In response to the pandemic, our UK NHS Trust isolated the smallest of four hospital sites and re-designated it as an elective surgical COVID-secure surgical facility. We aimed to analyse the outcomes for patients who had major surgery at this site in the first five months since its inception, in particular in relation to their perioperative management within this context.

Materials and methods

Study design

A retrospective observational study was undertaken to include all consecutive patients who underwent major surgery between its inception on 8 June 2020 and 30 November 2020 at a newly established COVID-secure facility. Major surgery was defined as patients requiring admission to the level 2 Post Anaesthetic Care Unit (PACU) postoperatively rather than going directly to a level 1 ward bed.

Study setting

Our NHS Trust is one of the largest NHS Trusts in England and consists of four hospital sites. The largest hospital is one of the busiest COVID Intensive Care Units (ICU) units in Europe. To ensure continuation of elective surgical care, the smallest site within the Trust was converted to an elective COVID-secure surgical facility. There are seven operating theatres, a PACU and three surgical wards in addition to an admissions unit. The PACU can offer level 2 care to patients in the immediate postoperative period, with a pathway designed to discharge patients back to the ward within 23h of admission to PACU. If patients require prolonged level 2 care, then they are transferred to one of the other acute non-COVID-secure hospital sites within the NHS Trust. Patients were required to have a COVID-19 swab remotely three and one days prior to admission and to then self-isolate from the time of their first swab until admission. Institutional approval for this study was granted prior to data collection as part of a Trust Quality Improvement project.

Patient selection

All patients who underwent major surgery (defined as admission to PACU postoperatively) were included, from the general surgery, urology or otorhinolaryngology (ENT) services. Patients were excluded if their operation was cancelled, they had anaesthetic complications or other pathology deeming them unfit for surgery.

Data collection

Patient data included age, gender, American Society of Anesthesiologists (ASA) grade, malignancy status, type of neuraxial anaesthesia (ie: epidural or spinal), PACU admission duration and transfer to another unit within the NHS Trust. Study outcomes included 30-day complications, readmission, unplanned return to theatre, COVID-19 infection and length of stay.

Data analysis

Summary data are presented as n (%) for categorical data or median and interquartile range (IQR) for continuous data. Binary logistic regression and linear regression models were used to analyse the odds ratio (OR) and 95% confidence interval (95% CI) for categorical and continuous outcomes, respectively. Independent co-variable of interest were chosen a priori as age, gender, malignancy status and ASA grade. A p-value of <0.05 was deemed statistically significant.

Results

Patient characteristics

There were 282 patients identified during the study period. Two patients were excluded because they did not proceed to planned surgery. The first was abandoned due to anaphylaxis at induction of anaesthesia and the second due to a suspected cardiac event, making our study cohort 280 patients. Patient characteristics are summarised in Table 1. There were 159 (56.7%) male patients. The median age was 63.5 (IQR 52–75) years. The median ASA was 2 (IQR 2–3). 167 (59.6%) operations were undertaken for malignant pathology. For those who had neuroaxial anaesthesia, 16 (5.7%) patients had epidural anaesthesia and 115 (41.1%) had spinal anaesthesia. The choice of neuroaxial anaesthesia was made by the consultant anaesthetist responsible for delivery of the perioperative care of the patient on the day.

Table 1.

Study patient characteristics

Patient characteristics	All (n = 280)
Age, median (IQR)	63.5 (52–75)
Male, n (%)	159 (56.7)
ASA, median (IQR)	2 (2–3)
Malignant pathology, n (%)	167 (59.6)
PACU >23h, n (%)	55 (19.6)
Length of stay, median (IQR)	3 (2–6)
Transfer to another hospital, n (%)	24 (8.6)
30-day complication, n (%)	111 (39.6)
30-day readmission, n (%)	40 (14.3)
30-day reoperation, n (%)	13 (4.6)
30-day mortality, n (%)	1 (0.4)

IQR: interquartile range; ASA: American Society of Anesthesiologists; PACU: Post Anaesthetic Care Unit.

Utilisation of the PACU

There were 224 (80%) patients who were discharged from PACU in less than 23h according to the standard operating procedure for the Trust. There were 24 (8.6%) patients who were transferred to other acute (non-COVID-secure) sites within the NHS Trust. Ten (41.6%) of these 24 patients were transferred to level 2 or 3 care. PACU admission for >23h was associated with a higher chance of 30-day complications (OR 2.3 (95% CI 1.2, 4.3); p = 0.011) but not 30-day readmission or reoperation or hospital transfer (Table 2). Only age was a significant determinate of 30-day readmission (OR 0.97 (95% CI 0.95, 0.99); p = 0.015) (Table 2).

Table 2.

Summary of binomial logistic regression models to determine the OR of 30-day outcomes and hospital transfer according to age, gender, malignancy status, ASA grade, length of stay in PACU and use of perioperative epidural

Independent variable	30-day complicationsOR (95% CI); p-value	30-day reoperationOR (95% CI); p-value	30-day readmissionOR (95% CI); p-value	Hospital transferOR (95% CI); p-value
Age	1.00 (0.99, 1.02); 0.74	0.96 (0.93, 0.999); 0.06	0.97 (0.95, 0.99); 0.01a	1.02 (0.98, 1.05); 0.325
Male	1.11 (0.66, 1.87); 0.69	2.9 (0.84, 13.46); 0.12	1.2 (0.599, 2.47); 0.61	0.98 (0.41, 2.45); 0.970
Malignancy	0.67 (0.39, 1.17); 0.16	0.8 (0.21, 3.07), 0.74	1.42 (0.66, 3.19); 0.38	1.19 (0.45, 3.40); 0.738
ASA	0.93 (0.59, 1.45); 0.74	1.44 (0.54, 3.68); 0.45	0.97 (0.52, 1.78); 0.93	0.92 (0.42, 1.95); 0.829
PACU >23h	2.28 (1.21, 4.32); 0.01a	1.51 (0.30, 5.76); 0.57	1.08 (0.42, 2.51); 0.86	2.19 (0.82, 5.48); 0.102
Epidural	4.91 (1.58, 18.55); 0.01a	1.29 (0.06, 9.02); 0.83	1.08 (0.42, 1.51); 0.72	4.32 (1.19, 14.2); 0.003a

Note: ASA: American Society of Anesthesiologists; PACU: Post Anaesthetic Care Unit; OR: odds ratio; 95% CI: 95% confidence interval.

^aStatistically significant.

Complications

There were 111 (39.6%) patients who had a complication within 30 days, with a median Clavien-Dindo score of 2 (IQR 2–2). Forty (14.3%) patients were readmitted within 30 days; 18/40 (45%) of these readmissions were discharged within 24h. Thirteen (4.6%) patients had a reoperation within 30 days, 6/13 (46.2%) of these patients had been transferred to a non-COVID-secure site on their primary admission. There was one (0.4%) mortality within 30 days of operation.

Neuroaxial anaesthesia

Having neuroaxial anaesthesia (either epidural or spinal) was associated with an increased length of stay (OR 18.8 (95% CI 4.4, 80.9); p < 0.001) and increased risk of complications (OR 3.2 (95% CI 1.9, 5.5); p < 0.001), but there was no effect on the likelihood of a PACU stay of >23h (p = 0.067), 30-day reoperation rate (p = 0.17) or readmission (p = 0.07) (Table 3). Epidural use was associated with increased length of stay (OR 118 (95% CI 5.5, 2532); p = 0.0024), lower chance of leaving PACU within 23h (OR 0.22 (95% CI 0.08, 0.65); p = 0.006), increased risk of 30-day complications (OR 4.3 (95% CI 1.5, 14.3); p = 0.009) and increased chance of transfer to a hot site (OR 4.3 (95% CI 1.2, 14.2); p = 0.019). Epidural use was not predictive of 30-day readmission (0.61), but age was (OR 1.03 (95% CI 1.01, 1.06); p = 0.015) (Table 4).

Table 3.

Summary of linear regression and binomial logistic regression models to determine the OR of 30-day outcomes, length of stay and PACU admission breach according to age, gender, malignancy status, ASA grade and use of any neuroaxial anaesthesia

Independent variable	30-day complications OR (95% CI); p-value	30-day reoperation OR (95% CI); p-value	30-day readmission OR (95% CI); p-value	Length of stay (95% CI); p-value
Age	1.01 (0.99, 1.02); 0.46	0.96 (092, 1.0); 0.06	0.97 (0.09, 2.99), 0.01a	1.04 (0.99,1.09); 0.091
Male	0.98 (0.58, 1.65); 0.94	2.698 (0.78, 12.49); 0.15	1.16 (0.58, 2.37); 0.69	1.15 (0.27, 4.82); 0.85
Malignancy	0.71 (0.39, 1.26); 0.24	0.699 (0.18, 2.79); 0.60	1.28 (0.58, 2.90); 0.55	1.43 (0.296, 6.86); 0.66
ASA	1.31 (0.82, 2.09); 0.26	1.67 (0.59, 4.61); 0.32	1.05 (0.55, 1.96); 0.88	1.61 (0.46, 5.69); 0.46
Epidural or spinal	3.17 (1.86, 5.51); <0.001a	2.36 (0.71, 8.83); 0.17	1.94 (0.95, 4.06); 0.07	18.81 (4.37, 80.90); <0.001a

Note: ASA: American Society of Anesthesiologists; PACU: Post Anaesthetic Care Unit; OR: odds ratio; 95% CI: 95% confidence interval.

^aStatistically significant.

Table 4.

Summary of linear regression and binomial logistic regression models to determine the OR of 30-day outcomes, length of stay and PACU admission breach according to age, gender, malignancy status, ASA grade and epidural anaesthesia

Independent variable	30-day complications OR (95% CI); p-value	30-day reoperation OR (95% CI); p-value	30-day readmission OR (95% CI); p-value	Length of stay (95% CI); p-value
Age	1.01 (0.59, 1.65); 0.43	0.96 (0.93, 1.0); 0.06	1.03 (1.01, 1.06); 0.02a	1.04 (0.99, 1.10); 0.08
Male	0.99 (0.59, 1.65); 0.96	2.85 (0.83, 13.16); 0.13	0.84 (0.41, 1.67); 0.62	1.18 (0.28, 5.02); 0.83
Malignancy	0.86 (0.497, 1.49); 0.59	0.81 (0.22, 3.13); 0.75	0.70 (0.21, 3.26); 0.37	2.11 (0.44, 10.04); 0.35
ASA	1.1 (0.70, 1.72); 0.67	1.46 (0.54, 3.74); 0.44	1.02 (0.56, 1.91);	1.09 (0.31, 3.87); 0.89
Epidural	4.34 (1.51, 14.31); 0.009a	1.54 (0.08, 9.57); 0.696	0.71 (0.21, 3.26); 0.61	118.16 (5.51, 2532.96); 0.002a

Note: ASA: American Society of Anesthesiologists; PACU: Post Anaesthetic Care Unit; OR: odds ratio; 95% CI: 95% confidence interval.

^aStatistically significant.

Nosocomial COVID-19

Two patients were diagnosed with nosocomial COVID-19. Both patients were transferred to an acute (non-COVID-secure) site following their operation. The first had bilateral ureteric injuries associated with placement of ureteric stents prior to an anterior resection. The patient required nephrostomies, which could only be inserted at the acute site. The second was transferred to an acute site due to bowel obstruction caused by a Richter’s hernia following a redo anterior resection and resection of seminal vesicles for recurrent rectal cancer. Neither required admission to ICU for COVID-19. Both patients had their primary operation in November 2020.

Discussion

We report the utilisation of a COVID-secure site for elective surgery for the delivery of safe surgical care to patients during a viral pandemic. Thirty-day mortality of all major operations was low at 0.4%. Postoperative care for patients following major surgery was undertaken using a PACU, and those who required a PACU stay of over 23h were more likely to have 30-day complications. Both spinal and epidural anaesthesia were associated with increased length of stay and 30-day complications, although it is possible that this may be a surrogate marker of patients undergoing more extensive or major surgery. Epidural use was associated with increased length of stay, PACU stay longer than 23h, 30-day complications and increased risk of transfer to a non-COVID-secure site.

There were two patients (0.7%) in our series who contracted COVID-19 within 30 days of surgery, after transfer to a non-COVID-secure site. Neither had complications from a COVID-19 perspective and neither required invasive ventilation. Both cases occurred in November 2020, when there was an increasing incidence among the local community compared to when the elective surgical site was established at the start of the study.

Perioperative SARS-CoV-2 infection is associated with postoperative pulmonary complications in 51.2% and 30-day mortality of 23.8% (Nepogodiev et al 2020). Due to level 2/3 bed availability and staff redeployments during the first wave of COVID-19, non-elective surgical procedures were cancelled. Now, as elective surgery resumes, data show that COVID-19 free pathways can reduce the pulmonary complications and SARS-CoV-2 infection rate (Chang et al 2020a, Glasbey et al 2021). To ensure biosecurity and minimise transmission, it is imperative that patients and staff are tested to ensure no asymptomatic carriage and transmission within the hospital environment. The Royal College of Surgeons of England (2020) recommended staff be tested twice a week and patients prior to admission and discharge home. Some countries in East Asia are testing staff daily to prevent asymptomatic transmission. Although individual hospital pathways will vary dependant on the Trust’s configuration, the key is to reduce transmission by cohorting patients and staff, personal protective equipment, excluding visitors from hospital sites, regular testing and expedited vaccination for patients and staff. The model of care that we report in this study facilitates the cohorting of staff and patients more effectively and may prevent the potential for infection when there are COVID-19 patients situated in the same site. The recent COVIDSurg study (Li et al 2020) reported that postoperative SARS-CoV-2 infection is associated with an increased mortality and therefore enabling these measures may help to prevent such a risk to patients.

To further minimise risk to patients and optimise outcomes, the length of stay in hospital should be minimised to the shortest safe duration. Data from our study show that epidural anaesthesia was associated with an increased length of stay, longer stay in PACU, 30-day complications and increased risk of transfer to a non-COVID site, independent of age, gender, malignancy status and ASA grade. Both epidural and spinal anaesthetic are shown to provide superior analgesia for elective laparoscopic colorectal surgery than patient-controlled analgesia (PCA) alone (Brown et al 2020). Levy et al (2011) provided randomised data demonstrating epidural to be significantly worse than spinal analgesia or PCA in terms of length of stay and return of bowel function in laparoscopic colorectal surgery. Additionally, they showed no significant difference between pain scores of patients randomised to epidural vs spinal anaesthesia. There are well-recognised side effects of epidurals including hypotension, urinary retention and pruritis (Garimella & Cellini 2013, Marret et al 2007). The former of these may lead to excess intravenous fluid infiltration which may contribute to ileus and therefore delayed recovery (Levy et al 2011). Epidurals also require skilled nursing care in the postoperative period which spinal anaesthesia do not (Garimella & Cellini 2013, Marret et al 2007). This is an important consideration when critical care staff numbers are limited and may be redeployed during a pandemic.

The risk of prolonged hospital admission in the COVID era also emphasises the importance of adoption of Enhanced Recovery After Surgery (ERAS) principles, and a purely elective hospital environment allows reinforcement of protocol-based care without the additional logistic burden of emergency admissions.

Limitations

The data from this study are retrospective, with all of the limitations and risk of bias associated with that design. Only patients with positive COVID-19 swabs were considered to have acquired nosocomial SARS-CoV-2, but we acknowledge the possibility that there may have been some false negatives that were not picked up through testing. There may have also been some patients with asymptomatic or community SARS-CoV-2 infection that did not require admission to hospital and were not tested. However, given the primary concern is regarding increased postoperative mortality and pulmonary complications, we are reassured that the lack of admissions with COVID positive status likely indicates relative success of the pathway. Although all included patients were considered to have had major surgery, those with epidural anaesthesia may have had larger or more lengthy surgery, which may have introduced some selection bias when analysing outcomes. We did not include patient procedures from a similar time period prior to the pandemic, so direct comparison with pre-COVID patients was not possible. However, there were likely to be many differences in patient pathways and staffing between the study period and the pre-COVID period, so a comparison may have been at risk of multiple biases.

The future

It is predicted there will be over 3000 additional deaths in five years and approximately 60,000 years of life lost due to diagnostic delays in breast, colorectal, lung and oesophageal cancer (Maringe et al 2020). Chang et al (2020a) showed only 56% of their orthopaedic patients agreed to proceed to surgery due to hesitancy regarding nosocomial COVID-19, and therefore, we must demonstrate our surgical pathways to be safe. Ten million people are now awaiting a surgical procedure in the UK, and therefore, we must develop pathways to deliver consistent surgical care to not only meet the influx of new referrals, but to also clear the backlog. Without timely surgery, patients are more likely to present as emergencies or with complications, which will increase resource cost (Sud et al 2020). We should harness the principles of ERAS, optimise perioperative anaesthesia and attempt to minimise hospital length of stay for patients. Restructuring sites may facilitate safer elective surgery.

Conclusion

Establishing a COVID-secure site enabled continuation of major urgent elective surgery within a viral pandemic. Level 2 support is essential to ensure safe delivery of complex surgery at ‘cold’ (COVID-secure) sites. From our initial data, epidural anaesthesia may not be the ideal neuroaxial anaesthesia for this context, but further prospective investigations are warranted. Optimal perioperative anaesthesia and enhanced recovery pathways may be the key to reducing transfer to non-COVID-secure sites and reducing nosocomial infection.

No competing interests declared.