Appendicitis in the COVID-19 era: a modern challenge for experienced hands

Abstract

Introduction

The first wave of COVID-19 was accompanied by global uncertainty. Delayed presentation of patients to hospitals ensued, with surgical pathologies no exception. This study aimed to assess whether delayed presentations resulted in more complex appendicectomies during the first wave of COVID-19.

Methods

Operation notes for all presentations of appendicitis (n=216) within a single health board (three hospitals) during two three-month periods (control period (pre-COVID) vs COVID pandemic) were analysed, and the severity of appendicitis was recorded as per the American Association for the Surgery of Trauma (AAST) grading system.

Results

Presentations of appendicitis were delayed during the COVID period with a median duration of symptoms prior to hospital attendance of two days versus one day (p=0.003) with individuals presenting with higher median white cell count than during the control period (14.9 vs 13.3, p=0.031). Use of preoperative CT scanning (OR 3.013, 95% CI 1.694–5.358, p<0.001) increased significantly. More complex appendicectomies (AAST grade >1) were performed (OR 2.102, 95% CI 1.155–3.826, p=0.015) with a greater consultant presence during operations (OR 4.740, 95% CI 2.523–8.903, p<0.001). Despite the greater AAST scores recorded during the COVID period, no increase in postoperative complications was observed (OR 1.145, 95% CI 0.404–3.244, p=0.798).

Conclusions

Delayed presentations during the COVID-19 pandemic were associated with more complex cases of appendicitis. Important lessons can be learnt from the changes in practice employed as a result of this global pandemic.

Introduction

The first case of COVID-19 in the UK was identified in January 2020. By March 2020 most areas of the UK had positive cases with local hospitals and intensive care units impacted by the virus that had taken the world by storm.¹ The virus affected all areas of the medical community including surgical departments, where elective work, including cancer operations, were stopped in order to enable hospitals to focus on and accommodate the influx of COVID-19-positive patients. An unforeseen consequence of the rapid spread of the virus was a reduction in emergency presentations across the entire spectrum of surgical pathologies. Patel et al recently identified that there was also delayed presentation of patients with such pathologies. Interestingly, despite this delay, no significant difference in the patients’ physiological or biochemical markers on admission were identified when compared to a pre-lockdown cohort.²

Acute appendicitis is the most common abdominal surgical emergency globally with approximately 50,000 appendicectomies performed per year in the UK alone.³ Prior to the COVID-19 pandemic, appendicitis remained primarily a clinical diagnosis with use of radiological investigations variable among countries, health boards and even individual centres. However, during the COVID-19 pandemic, the World Health Organisation (WHO) and the surgical Royal Colleges in the UK advocated changes in the investigation and management of surgical presentations including suspected appendicitis. Increased use of CT scanning in augmenting diagnoses and the use of antibiotics as first line management in early appendicitis were recommended. Owing to concerns surrounding laparoscopy being an aerosol-generating procedure, open appendicectomy was deemed the optimal approach. These changes sit outside the ‘normal’ practice in the UK where early laparoscopic appendicectomy is the gold standard.⁴

The American Association for the Surgery of Trauma (AAST) grading system for anatomic severity of appendicitis has been validated and grades the severity of appendicitis using multiple criteria.^5–7 This study aimed to assess whether delayed presentation of appendicitis resulted in more complex appendicectomies during the COVID-19 first wave using the AAST operative grading criteria.

Methods

A multi-site retrospective analysis was performed across three sites within a single University Health Board. Patients were identified via the health board’s coding department, where a search was performed for all patients discharged with the diagnosis of appendicitis within the two study periods. To ensure that no patients who met the inclusion criteria were missed, inpatient lists were searched for patients who presented with right iliac fossa pain or who were diagnosed with suspected appendicitis on admission. Data collected related to patient demographics, length of stay, duration of symptoms prior to attendance, biochemical/haematological results, preoperative imaging, COVID status, surgical versus conservative management, surgical approach, operative details/findings (where applicable) and postoperative management including complications. Data were collected using the Welsh Clinical Portal (blood, radiology and histology results, scanned clinical and operation notes, discharge summaries and follow-up clinic letters) and where data was missing, the case notes were requested and subsequently analysed. Ethical approval was deemed not to be required by the decision panel of the Health Research Authority as the work represents clinical surveillance/service evaluation.

Rationale for study period

Data pertaining to patients diagnosed with appendicitis were collected for two study periods. The COVID period ran from 16 March 2020 (when the first patient was admitted to the largest hospital in the health board) to 16 June 2020 (when the emergency surgical take had returned to normal capacity). The control study period (pre-COVID) ran from 1 November 2020 to 1 February 2020, ending before the pandemic spread aggressively through the UK and prior to the first COVID-19 positive admission to hospital nationally.

Inclusion/exclusion criteria

All patients with a discharge diagnosis of appendicitis were included in the study. Patients were excluded if they were discharged with an alternative diagnosis or radiological reports or operative findings suggested an alternative diagnosis. Owing to delays in histology reports, patients are more often than not discharged prior to histological confirmation of appendicitis, and therefore the discharge diagnosis was based on clinical and radiological findings. Histology was followed up and a negative appendicectomy rate reported for each cohort.

The American Association for the Surgery of Trauma grading system

The AAST have developed a wide array of severity grading scores for emergency general surgical conditions, which are broken down into four different criteria: clinical, imaging, operative and pathological. A retrospective analysis of operation notes and radiology reports (CT scans only as per the AAST guidelines) was performed and a severity grade was assigned to each operation/CT scan based on the description of the appendix (and associated findings) as described by the primary operating surgeon/reporting radiologist. Table 1 illustrates the anatomic severity system as defined by the AAST.^8,9

Table 1 .

The American Association for the Surgery of Trauma (AAST) anatomic severity table for acute appendicitis^8,9

AAST grade	Description	Imaging (CT) criteria	Operative criteria
1	Acutely inflamed	Inflammatory changes confined to the appendix ± dilated ± contrast non-filling	Acutely inflamed appendix
2	Gangrenous	Appendiceal wall necrosis with contrast non-enhancement ± air in the appendiceal wall	Gangrenous (intact) appendix
3	Perforated with local contamination	Localised fluid ± contrast extravasation	Perforated appendix with evidence of local contamination
4	Perforated with periappendiceal phlegmon or abscess	Regional soft tissue inflammatory changes, phlegmon or abscess	Perforated appendix with associated abscess of phlegmon in the region of the appendix
5	Perforated with generalised peritonitis	Diffuse abdominal or pelvic inflammatory changes ± free intraperitoneal air or fluid	Perforated appendix with generalised purulent contamination away from the appendix

Justification for sample size

Sample size calculations were based on a pre-study literature analysis which highlighted that gangrenous (AAST grade 2) and perforated appendicitis (AAST 3+) accounts for approximately 30% of all appendicectomies, and an enrolment ratio of 0.5 was in line with those reported in a previous study.^2,4,5,10 A 20% increase in the prevalence of these more complicated cases (AAST >1) was deemed as an appropriate clinically relevant effect size. Therefore, a minimum of 205 patients were to be studied, providing 80% power to detect such a difference with p<0.05.

Statistical analysis

A Shapiro–Wilk test was performed, which identified the data as not normally distributed (p<0.05). Therefore, statistical analysis appropriate for non-parametric data was performed using SPSS Statistics (Version 25.0; IBM Corp., Armonk, NY, USA). Mann–Whitney U-test and binary logistic regression were used to compare variables between the pre-COVID and COVID period. To assess agreement between AAST grade derived from CT scan reports and operative findings, Kappa statistic (k) and Altman’s rating of strength of agreement were used: k<0.20 (poor), k=0.21–0.40 (fair), k=0.41–0.60 (moderate), 0.61–0.80 (good), 0.81–1.00 (excellent).^11,12

Results

A total of 216 patients were diagnosed with appendicitis across the two study periods (138 pre-COVID vs 78 COVID period, 43.5% reduction). The overall, pre-COVID and COVID period median age at presentation was 29 (IQR 18–41), 30 (17–45.25) and 28 (18–39) years old, respectively (p=0.411). The study included 113 male and 103 female patients with no significant gender variation between the two study periods (M:F, 70:68 vs 43:35, p=0.534).

Preoperative investigations and findings

The median time to presentation following the onset of symptoms was delayed during the COVID period compared with the pre-COVID period (2 (IQR 1–4) days vs 1 (1–2) day, p=0.002). Individuals presenting during the COVID period were more likely to undergo radiological imaging (OR 1.782, 95% CI 1.004–3.163, p=0.048); however, the pattern of modality use changed from predominantly ultrasound scans (OR 0.327, 95% CI 0.129–0.830, p=0.019) to CT scans (OR 3.013, 95% CI 1.694–5.358, p<0.001). Further variations in the preoperative investigations and findings can be found in Table 2. Despite a total of 78 diagnoses of appendicitis being made during the COVID period, only 11 (14%) individuals had a COVID swab (one positive, ten negative).

Table 2 .

Patient demographics across the two study periods

	Pre-COVID (n=138)	COVID (n=78)	p-value
White cell count	13.3 (9.6–16.9)	14.9 (11.9–17.7)	0.031
C-reactive protein	32 (7.0–78.5)	31 (9.0–99.5)	0.622
AKI present	3 (2.2%)	1 (1.3%)	0.635
Radiological imaging
None	67 (48.6%)	27 (34.6%)	0.048
Ultrasound	28 (20.3%)	6 (7.7%)	0.019
CT	43 (31.1%)	45 (57.7%)	<0.001
Radiological AAST grade*			0.708
1	23 (53.5%)	23 (51.1%)
2	2 (4.6%)	3 (6.7%)
3	6 (14.0%)	6 (13.3%)
4	8 (18.6%)	6 (13.3%)
5	1 (2.3%)	4 (8.9%)
Negative	3 (7%)	3 (6.7%)

Figures are medians or cohort population number with interquartile ranges or percentages in parentheses

AAST = American Association for the Surgery of Trauma; AKI = acute kidney injury; COVID = coronavirus disease; CT = computerised tomography

*Radiological AAST grade is only applicable to CT scans

Operation details and findings

Thirteen patients were treated conservatively with antibiotic therapy (1 pre-COVID, 12 COVID period). Of those undergoing an appendicectomy, a consultant surgeon was more likely to be present (OR 4.740, 95% CI 2.523–8.903, p<0.001), an open approach employed (OR 11.611, 95% CI 5.761–23.400, p<0.001) and individuals were more likely to have a gangrenous or perforated appendix (AAST≥2, OR 2.102, 95% CI 1.155–3.826, p=0.015) during the COVID period compared with the pre-COVID period. Operative AAST ≥2 was associated with a longer median duration of symptoms prior to presentation (2 (IQR 1–4) vs 1 (1–3), p=0.022) across the whole cohort. Table 3 illustrates the operative variation between the two study periods.

Table 3 .

Operative variations across the two study periods

	Pre-COVID (n=138)	COVID (n=78)	p-value
Surgical intervention	137 (99.3%)	66 (84.6%)	0.002
Consultant present	32 (23.4%)	39 (59.1%)	<0.001
Primary surgical approach			<0.001
Laparoscopic	118 (86.1%)	23 (34.8%)
Open	19 (13.9%)	43 (65.2%)
Operative AAST grade			0.042
1	83 (60.6%)	30 (45.5%)
2	8 (5.9%)	8 (12.1%)
3	15 (10.9%)	10 (15.2%)
4	14 (10.2%)	9 (13.6%)
5	9 (6.6%)	7 (10.6%)
Negative*	8 (5.8%)	2 (3.0%)
Negative appendicectomy (histology)	23 (16.8%)	7 (10.6%)	0.249

Figures are cohort population numbers with percentages in parentheses

AAST = American Association for the Surgery of Trauma; COVID = coronavirus disease; CT = computerised tomography

*Negative appendicectomy in relation to operative AAST is based on the primary operator’s reported findings during the procedure; the true negative appendicectomy rate is based on the histological report

AAST grade agreement between CT scan reports and operative findings

For individuals who underwent a preoperative CT scan, the agreement between the radiological and operative AAST grade was moderate (57.9% agreement, k=0.41) with CT reporting lower and higher AAST grades than reported operatively in 32.9% and 9.2% of individuals, respectively.

Although the AAST scoring system is only applicable to CT scans, this study also assessed the relative efficacy of ultrasound scanning. Thirty-four individuals underwent a preoperative ultrasound, of which 26 reported either that appendicitis could not be ruled out using ultrasound alone or only commented on pelvic gynaecological organs. Of these 26 individuals, 25 went on to have an appendicectomy with 15 individuals having histologically confirmed appendicitis.

Postoperative care

During the COVID period, antibiotics were more likely to be given postoperatively (72.7% vs 55.5%, OR 2.140, 95% CI 1.131–4.051, p=0.019). However, when given, there was no statistical difference in the median duration of the course when compared with the pre-COVID period (2 (IQR 1–5) vs 5 (1–7) days, p=0.163). No difference was identified relating to the incidence of postoperative complications between the pre-COVID and COVID cohorts (8.0% vs 9.1%, OR 1.145, 95% CI 0.404–3.244, p=0.798); however, significantly more complications were identified in individuals with an operative AAST ≥2 across the whole study population (13.8% vs 4.9%, OR 3.109, 95% CI 1.101–8.780, p=0.032). No deaths were recorded in the cohort in either study period. Despite the delayed presentation and increased severity of operative findings associated with the COVID period, there was no difference in the median length of stay (2 (1–3) vs 2 (1–3) days, p=0.545). Only one of the 12 individuals who were treated conservatively with antibiotics alone returned within the three-month COVID study period for failed management and required an appendicectomy. None of the other individuals who underwent appendicectomy had previously been treated with conservative management. There were no COVID-19-related postoperative respiratory complications identified.

Discussion

The principal findings of this study were that statistically significant delays in presentation were identified during the COVID period when compared with the control period. Associated reporting of more complex intraoperative findings was observed, with appendicitis of AAST grade 2 or more twice as likely. Greater use of preoperative imaging was also noted with CT scans three times more likely to be performed than during the pre-COVID period. This increase in CT scanning is also likely to account for the minimal number of COVID swabs performed, as CT scans were also utilised as COVID screening tools as initially advised by the Royal Colleges of the UK.¹³ Variations in operative practice were detected, with an almost 5-fold increase in consultant presence during operations and 12-fold increase in the primary surgical approach being an open operation. These changes in practice align with the recommendations made by WHO and the UK Royal Colleges as previously outlined. However, although the uptake of non-operative management as a primary measure was greater than the control period, operative intervention was still the mainstay of treatment throughout the first wave of the pandemic in the studied population.

The findings of this study align well with those previously reported. Patel et al examined the frequency and severity of general surgical emergency admissions during the COVID-19 pandemic.² They identified a 50.3% reduction in presentations during the lockdown period when compared to the same time period a year prior, with this study similarly identifying a 43.5% reduction in presentations. Moreover, the number of days of symptoms prior to presentation was significantly higher during lockdown (four vs three days), a similar pattern to that seen in this study. However, Patel et al reported that patients were not more unwell despite this delayed presentation, whereas this study identified that patients had higher median white cell counts and a greater operator perceived complexity (AAST grade) in relation to the surgical findings.

In 2017, Matthew Hernandez et al performed validation studies of the AAST grading system in both developed and developing nations (US and South Africa). Higher AAST grades were shown to be associated with greater postoperative complication rates in both populations.^5,6 Similarly, in this study, higher operative AAST grades were associated with a greater than three-fold incidence of postoperative complications. Interestingly, however, the current study did not demonstrate a greater incidence of postoperative complications during the COVID period despite the significantly higher AAST grades being identified. This could be due to partial mitigation through the increased consultant presence during operative intervention and the increased utilisation of preoperative imaging in relation to operative planning.

The World Society of Emergency Surgery (WSES) has recently published the updated Jerusalem guidelines.¹⁴ The key recommendations surrounding preoperative work-up advocate the use of risk stratification models similar to those recommended by the Right Iliac Fossa Treatment (RIFT) study.¹⁵ Subsequent utilisation of ultrasound scanning, performed by an experienced operator, is deemed a reliable initial investigation followed by low-dose contrast CT scan where further investigation is indicated. Owing to the risk of occult malignancy in older populations, it is widely accepted that a CT scan should be performed as a primary investigation in those over 50 years old.¹⁶ The increased utilisation of CT scans and the greater than one-third reduction in the histologically proven negative appendicectomy rate reported during the COVID period, as outlined in this study, would advocate a change in current practice, not just during the pandemic but also in general terms, to align with those recommendations made by the Jerusalem guidelines. However, the reported efficacy of ultrasound in this study is significantly less than that outlined in the evidence supporting its use in the Jerusalem guidelines; therefore, this study would recommend that ultrasound scanning is reserved for individuals who are most likely to gain an alternative diagnosis for their pain, such as those most at risk of gynaecological pathology.

The early move towards open appendicectomies and the increased presence of consultant surgeons is a novel finding and offers opportunity beyond just that of patient safety. The cessation of elective work during the pandemic resulted in a significant reduction in training opportunities. Equally, despite the modern surgical registrar being well versed in the laparoscopic approach to appendicectomy, their experience of open procedures is often a fraction of that of their predecessors, and as such, this offered an excellent opportunity for training. Observationally, although there was a five-fold increase in the presence of consultant surgeons, this was primarily as second surgeon and therefore suggests that active training in the emergency setting was maximised. These experiences will no doubt better equip the current cohort of trainee surgeons with new skills and leaves food for thought regarding future maximisation of training opportunities even outside this global pandemic.

Study limitations

This study has a number of inherent limitations. Firstly, patients were from a single geographical area of the UK, and therefore the findings of this study would be strengthened by a larger scale study encompassing other UK regions or by obtaining a global sample/perspective. Despite the multiple approaches used to minimise the number of patients missed during the two study periods, it is feasible that individuals may have been coded incorrectly during the administrative process or were not added to the patient lists and therefore may have been missed. This is most likely to affect individuals who were reviewed on an ambulatory basis and deemed either low risk for appendicitis or suitable candidates for a trial of conservative management. As they were never formally admitted to the hospital, the methodology employed would not capture this cohort. This, along with the possibility of changes in practice in primary care, may in part account for the significant reduction seen in presentations during the COVID period. In contrast, the study has several strengths, benefitting from extensive pre- and perioperative data sources allied with complete follow-up data. Where data were unavailable electronically, all case notes were accessible and the required data acquired. The grading system used by AAST has been validated in multiple settings and is therefore felt to be a reliable method for identifying operative complexity. Finally, the number of study participants exceeds the required population as outlined by the power calculation with no significant demographic variations between the two cohorts.

Conclusions

The findings of this study support the existing literature and the recommendations for practice advocated by the WHO and Royal Colleges during the COVID pandemic. Yet, these findings and the improvements in practice seen as a consequence of the pandemic align well with the most recent recommendations outlined by the Jerusalem guidelines and should prompt review of practice across the UK with the aim of reducing unnecessary admission and the significantly higher negative appendicectomy rate seen in the UK compared with other nations.¹⁵