The Impact of Surgical Delay: A Single Institutional Experience at the Epicenter of the COVID Pandemic Treatment Delays in Women with Endometrial Cancer and Endometrial Intraepithelial Hyperplasia

Abstract

The spread of COVID-19 led to a lockdown in New York in March of 2020. Nonemergent surgeries were postponed, including oncologic procedures. The backlog of surgeries was addressed starting May 2020. Our goal was to examine the change in waiting times for endometrial cancer surgeries during the COVID-19 pandemic in our institution. Data on surgery incidence and waiting time was gathered for patients diagnosed with endometrial intraepithelial neoplasia and endometrial cancer. The association between days from diagnosis to surgery was adjusted for age, obesity, presence of comorbid conditions, race, smoking history and diagnosis and was examined using a general linear model. A total of 190 patients were identified for this retrospective study. Five subjects were missing information on race and were excluded from all analyses, resulting in 185 subjects in the final analyses. Mean waiting time during COVID-19 was 70.9 days (95%CI 55.0, 91.3), compared to 49.3 (95%CI 49.8, 63.8) days during the reference period. No significant associations were seen between the time and any of the clinical or demographic factors.

1. Introduction

The global spread of coronavirus disease 2019 (COVID-19) following the first case reported in Hubei province, China, led to a nationwide lockdown in the United States during March of 2020. During this first wave, all nonemergent surgeries, including oncologic procedures, were postponed due to the impending patient volume to maintain intensive care bed capacity. By the summer of 2020, efforts were underway to address the accumulated backlog in all nonemergent surgeries [1,2].

An estimated 2.3 million cancer surgeries were delayed or cancelled during the first wave, with many patients reporting anxiety and concern for cancer progression [2]. The COVIDSurg gynecological cancer international study demonstrated that 11% of women undergoing surgical treatment experienced a delay in surgery [3]. Endometrial cancer is the most common gynecological malignancy in high-income countries, with 66,000 new cases diagnosed annually in the United States [4]. The precursor lesion of endometrioid adenocarcinoma is endometrial intraepithelial neoplasia (EIN), also known as atypical endometrial hyperplasia [5]. The standard treatment for endometrial cancer and EIN is hysterectomy with comprehensive surgical staging if indicated [6].

The main objective of this study was to evaluate changes in waiting times for surgical treatment of endometrial cancer (EC) and EIN that occurred after the initial lockdown ended in the summer of 2020. The secondary objective was to describe the characteristics and factors associated with treatment delay. We hypothesized that those with underlying medical conditions, as defined by the presence of cardiovascular disease and/or diabetes, and obesity would be more adversely affected by the surgical backlog that occurred after the initial lockdown ended.

2. Materials and Methods

This retrospective study was approved by the Institutional Review Board. All patients who underwent a hysterectomy for EIN or EC were identified at our institution from January 2019 through July 2021. Two observation periods were defined in relation to the mandated closure of the operating rooms to nonemergent cases during the first wave of the COVID-19 outbreak. At our institution, the operating rooms were closed by the end of March 2020 and reopened with minimal surgical bookings in June 2020. Period 1, or the reference period, encompassed the period for January 2019–March 2020, defined as usual care prior to the pandemic. Period 2 encompassed the dates from June 2020 to July 2021. The waiting time was calculated for all patients as the time from diagnosis by endometrial biopsy (via dilation and curettage or aspiration biopsy) and the date of the surgery. Patients were excluded if the time of diagnosis was unable to be obtained. Baseline characteristics were collected, including age, BMI, history of smoking, race/ethnicity, and presence of underlying comorbid conditions defined as diabetes, prediabetes, and cardiovascular disease. Patients were considered to have an upgrade of disease if, on final pathology after hysterectomy, there was a change in diagnosis from EIN to EC. Study data were managed using REDCap electronic data capture tools. A general linear model was used to analyze the effect that race/ethnicity, underlying medical conditions, smoking status, and obesity had on waiting times before and during the COVID era. The log of time was used to better meet the assumptions of the parametric model. Summary statistics are given as least squares means and their associated 95% confidence intervals calculated from the ANOVA model and then transformed back to the original unit of measurement (time in days).

3. Results

A total of 200 patients underwent a hysterectomy for EIN or EC at our institution from January 2019 through July 2021. Of the 200 patients, 10 were excluded due to missing data regarding the time of diagnosis. Demographic information for all patients meeting the study criteria is summarized in

Table 1. Demographic parameters of endometrial surgical oncology patients.

			Time Period
			Reference (Period 1)	COVID Era (Period 2)	Total
Age Mean (SD)			60 (10)	60 (11)	60 (10)
BMI Mean (SD)			36.9 (11.0)	36.2 (10.6)	36.21 (10.69)
Obesity	Non-obese	Count	50	35	85
		%	46.7%	44.9%	45.9%
	Obese	Count	57	43	100
		%	53.3%	55.1%	54.1%
Race/Ethnicity	White/Non-Hispanic	Count	55.1	49	108
		%	57.8%	62.8%	58.4%
	Black	Count	20	12	32
		%	18.7%	15.4%	17.3%
	Asian	Count	23	9	32
		%	21.5%	11.50%	17.3%
	Hispanic/Latino	Count	5	8	13
		%	4.7%	10.7%	7.0%
Comorbid Conditions	Yes	Count	59	34	93
		%	55.1%	43.9%	50.3%
	No	Count	48	44	92
		%	44.9%	56.4%	49.7%
Current Smoker	Yes	Count	6	4	10
		%	5.6%	5.1%	5.4%
	No	Count	101	74	175
		%	94.4%	94.9%	94.6%

.

The mean age at diagnosis was 60 (range 34–83). Ninety-six patients had underlying medical conditions, with 65.5% having cardiovascular disease, 15.6% diabetes or prediabetes, and 18.9% with multiple underlying conditions. Twelve patients were current smokers. Average BMI was 36.21 (SD 10.69). In total, 88 patients were non-obese (BMI < 35) and 102 patients were obese (BMI ≥ 35). A total of 61% of patients identified as White/Non-Hispanic, 18.1% Black, 16.4% Asian, and 4.5% Hispanic/Latino.

A total of 190 surgeries were included in the analysis. After the lockdown ended and elective surgeries resumed, the mean waiting time between diagnosis and surgical treatment increased compared to the reference period. The mean waiting time during COVID-19 was 70.9 days (95%CI 55.0, 91.3), compared to 49.3 (95%CI 49.8, 63.8) days during the reference period. This increase of 44% (48.3 days) was statistically significant (p < 0.00001). Using multivariant analysis, no significant associations were seen between the time and any of the clinical or demographic factors (

Table 2. Multivariable analysis evaluating clinical factors and surgical treatment delay.

	Level	Geometric Mean (95% CI)	p-Value
Time Period	Reference (Period 1)	49.3 (38.7, 62.9)	0.0004
	COVID era (Period 2)	70.9 (55.0, 91.3)
BMI	Non-obese	57.3 (44.4, 73.9)	0.5351
	Obese	61.0 (47.8, 77.9)
Comorbid Conditions	No	60.1 (47.0, 76.9)	0.7575
	Yes	58.2 (45.0, 75.1)
Race	Asian	46.0 (33.6, 62.9)	0.1006
	Black	69.7 (52.1, 93.3)
	Hispanic/Latino	63.4 (42.4, 94.9)
	White/Non-Hispanic	60.2 (47.7, 76.0)
Current Smoker	No	67.2 (58.1, 77.6)	0.2444
	Yes	52.1 (34.3, 79.1)
Diagnosis	Endometrial Cancer	52.5 (40.3, 68.3)	0.0262
	Endometrial Hyperplasia	66.6 (52.6, 84.5)

). Specifically, White–Non-Hispanic patients experienced a treatment delay of 60.2 days compared to 69.7 days amongst Black patients and 63.4 days for Hispanic patients. Irrespective of the time, the average surgical waiting time for women with EIN was longer compared to women with EC (p = 0.0134). During the reference period, women with EIN had an average waiting time of 60.1 days compared to 47.5 days for EC. After the lockdown ended, the average waiting time for EIN was 91.6 days compared to 68.5 days for women with EC. In addition, there was no statistically significant difference between the average BMI between the two time periods (p = 0.6525).

There was no association between time period and upgraded (p = 0.5852). In Period 1, 15.0% (16/107) of subjects were upgraded and, in Period 2, 17.9% (14/78) of subjects were upgraded.

4. Discussion

After the initial lockdown ended, our institution experienced a significant delay in the treatment of EH and EC. This delay may be due to multiple factors, including accumulation of backlogged cases during the cessation of nonemergent surgeries, new COVID-19 precautions for surgeries, resource shortages, and patient hesitation to schedule surgery during the pandemic. It is important to note we saw no association between race/ethnicity, underlying medical conditions, or obesity and surgical delay experienced after the lockdown ended. These results are in accordance with those reported in the literature. Specifically, Frey et al. reported that one third of patients with gynecologic cancer at three New York City hospitals experienced treatment delays, modifications, or cancellations during the COVID-19 pandemic. Of note, they found that race/ethnicity, primary cancer site, and the presence of medical comorbidities were not predictive of treatment modifications [7]. Another retrospective study found that, prior to the COVID pandemic, cancer treatment for African American patients was delayed by over 10 days compared to White patients. However, during the COVID-19 pandemic, this disparity was no longer present [8]. Similar results were also seen by Schmidt et al. This multicenter prospective cohort study demonstrated that, despite having higher rates of COVID infection, no difference in treatment delay was seen for Black patients compared to Non-Hispanic White patients. This was in contrast to Hispanic patients that showed to have a pandemic-related treatment delay [9]. The results of this study also demonstrated that, although there was a significant delay in surgical waiting time, the delay did not have an impact in the total number of patients that experienced an upgrade of disease from endometrial precancer to cancer.

There are some limitations to this study. First, we investigated the waiting times at a single institution rather than multiple centers across the country. While this limits the generalizability of these data, it also yields some key benefits. New York City was the epicenter of the COVID-19 lockdown in early 2020 and was more likely to suffer greater surgical delays than other locations around the country. Additionally, we did not investigate whether the surgical delay was due to the patients developing active COVID-19 infection between initial evaluation and scheduling of the surgery.

Well-differentiated EC and EIN are often considered an indolent disease with a favorable prognosis [10]. Therefore, a delay in surgical management of these conditions likely has little effect on disease progression or overall prognosis. However, it is important to note that a proportion of those women diagnosed with EIN on endometrial sampling will have an underlying high-grade EC pathology [11]. One retrospective study of women diagnosed with EIN demonstrated that 25.1% had concurrent EC and 3% had high-grade EC pathology. Thus, a delay could adversely affect those patients diagnosed with high-grade EC by allowing for disease progression and potentially increasing the risk of metastasis.

While the initial lockdown resulted in a significant delay in surgical treatment of women with EC and EIN, the COVID-19 pandemic has also had profound consequences in other areas of gynecological oncology. Recent publications have reported a decrease in gynecologic oncology practice volume of 61.6% since the start of the pandemic [12]. Additionally, studies have found between 19 and 35% fewer diagnoses of EC during the pandemic when compared to pre-pandemic rates [13,14]. While there has been a decrease in practice volumes due to provider-initiated cancellations as well as patient hesitancy to seek medical care, more advanced diseases have been noted at the time of initial presentation [15]. Future research will focus on identifying the potential additional consequences of the COVID-19 pandemic on women with EC, EH, and other gynecologic cancers.