ABSTRACT
INTRODUCTION: Prior literature suggests a need to delay elective surgery up to 8 weeks in patients with a previous SARS-CoV-2 infection to mitigate the risk of adverse postoperative events. However, these recommendations are broad without specific consideration of surgical procedures or factors related to the SARS-CoV-2 infection. The objective of this study was to identify optimal surgical timing in cancer patients with a prior history of COVID-19. METHOD(S): This study was a retrospective cross-sectional study using the National COVID Cohort Collaborative (N3C) centralized data resource. Patients who underwent select major surgeries after January 2020 were assessed for 30-day surgical outcomes. Cancer diagnoses, procedures, and outcomes were identified using standard SNOMED concepts. Patients with a history of COVID-19 prior to surgery were grouped by severity: non-hospitalized vs. hospitalized. Surgical timing relative to previous COVID-19 diagnosis was separated into three groups: 0-4 weeks, 4-8 weeks, and 8+ weeks. All analyses were performed using the N3C Data Enclave. RESULT(S): The study included 133,469 cancer patients with 7,757 (5.8%) who had a history of COVID-19 prior to surgery. Patients with prior COVID-19 did not have significantly increased rates of 30-day mortality (1.1% vs. 0.9%, P=.11) or non-fatal adverse events (23% vs. 22%, P=.33). Of the patients with a history of COVID-19, 1,630 (16%) were hospitalized due to a prior infection. Patients hospitalized for COVID-19 had significantly increased rates of postoperative 30-day mortality (2.9% vs. 0.5%, P< .001) and non-fatal adverse events (38% vs. 20%, P< .001) when compared to patients with a history of non-hospitalized COVID-19. Multivariable regression examining risk for any adverse event in patients with non-hospitalized COVID-19 demonstrated patients to be at baseline risk at 0-4 weeks (aOR 1.02 [0.68-1.50]), 4-8 weeks (aOR 0.98 [0.65-1.43]) and 8+ weeks (aOR 0.97 [0.80-1.17]) when compared to patients without a history of COVID-19. Conversely, patients previously hospitalized for COVID-19 were at increased odds of adverse postoperative events at all assessed time points (Figure). CONCLUSION(S): These data suggest prolonged delays from COVID-19 diagnosis to surgery are unnecessary among cancer patients not hospitalized for SARS-CoV-2 infection. However, there is a persistently elevated postoperative risk in patients who were hospitalized for treatment of COVID-19 prior to undergoing cancer surgery. Additional risk mitigation strategies beyond delaying surgery must be investigated.
ABSTRACT
INTRODUCTION The purpose of surveillance after resection of colorectal liver metastases (CLM) is to detect and treat recurrence using axial imaging, biomarker measurement, and a history/physical examination. In response to COVID-19 pandemic, telemedicine was used as a risk mitigation strategy to replace in-person visits, including for cancer surveillance. The objective of the study was to measure the uptake of telemedicine for cancer surveillance and outcomes following telemedicine surveillance after resection of CLM. METHODS Data from a prospective database was combined with real world data obtained from electronic health records using a cloud-based, data integration tool (Palantir Foundry) to identify patients in active surveillance following first surgical resection for CLM between April 2017 and April 2021. Telemedicine surveillance visit was defined as a follow-up visit >90 days following surgery using video or telephone. Recurrence was defined as detection of a new lesion. Bivariate statistical testing was performed using Student's t-test or chi-squared test. Retrospective chart review was used to validate identification of recurrence using the Foundry platform (100% interobserver agreement). RESULTS A total of 1,057 surveillance visits (306 patients) met our inclusion criteria. Prior to April 2020, 0% (0/686) visits utilized telemedicine. After April 2020, an average of 47.3% of visits per month utilized telemedicine (range 33.0 – 69.0%). The overall rate of identifying a recurrence during surveillance visit was 18.1% (191/1,057). There was no difference when comparing detection of recurrence using in-person (17.6%, 154/872) versus telemedicine visits (20.0%, 37/185, P=.371). The management of recurrence did not differ whether it was identified with an in-person or telemedicine visit;surgery, 36 (23%) vs. 10 (27%);ablation, 26 (17%) vs. 8 (22%);systemic therapy, 83 (54%) vs. 16 (43%);other, 9 (6%) vs. 3 (8%), respectively (P=.699). CONCLUSION Telemedicine was used in almost half of surveillance visits for CLM during the COVID- 19 pandemic. Detection and treatment of recurrence was similar for both telemedicine and in-person visits. Telemedicine-based follow-up is a safe and effective approach for surveillance after resection of CLM, supporting continued utilization beyond the pandemic.