Elective Surgery in the COVID-19 Era: A Screening Algorithm and Related Concerns

To the Editor:

Surgery in the era of coronavirus disease 2019 (COVID-19) is uncharted territory. Non-urgent surgery was discouraged at the beginning of the outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic. The need to reserve intensive care units for patients with COVID-19, the fear of spreading the infection through the hospital, and the possible worsening of surgical outcomes [1] were the main factors that motivated the postponement elective surgery. It has been estimated that as many as 28,404,603 operations were canceled during the first 12-week peak of the pandemic [2]. In the absence of scientific evidence on how best to resume surgical activity, we propose the algorithm used to screen potential surgical patients in a usually high-volume, tertiary referral university hospital in Rome, Italy (Fig. 1).

OPEN IN VIEWER

FIG. 1.

Flow chart of our screening algorithm. PCR = polymerase chain reaction; COVID-19 = coronavirus 2019.

Patients are screened by telephone interview for symptoms and recent contacts. If negative, patients access the hospital the morning of the day before surgery. Both rapid enzyme-linked immunosorbent assay (ELISA) for SARS-CoV-2 immunoglobulin G-immunoglobulin M (IgG-IgM) and reverse transcriptase-polymerase chain reaction (RT-PCR) for viral RNA in nasopharyngeal swabs are performed. The results of the serologic test are available within minutes. If negative, the patient is granted access to the ward with the RT-PCR result pending. Surgery is only confirmed for the next day once the swab result is negative, usually in the afternoon. If the ELISA rapid test is positive, the patient is sent to a special ward for suspected COVID-19 cases. In these circumstances, patients will only be confirmed for surgery if two consecutive RT-PCR swabs are negative. At any point if the RT-PCR swab is positive or symptoms develop, the patient is sent to a COVID-19 ward and the decision whether to perform surgery is then made by a multidisciplinary team.

Recently, a female patient with a liver recurrence from cervical squamous cell carcinoma was scheduled for an explorative laparotomy. She was found to be IgM-positive and was isolated immediately. Two consecutive RT-PCR results were negative, hence the patient proceeded to surgery 48 hours after admission. A segment 6 liver resection was performed. The operation and post-operative course were uneventful. Four months after surgery the patient is well. No hospital staff developed COVID-19 symptoms.

The viral RNA and ELISA rapid tests have different diagnostic windows, sensitivity, and specificity. The detection of viral RNA by RT-PCR is more likely from the onset of symptoms to three weeks afterward and has a specificity of 100% [3], whereas the detection of antibodies by ELISA likely becomes positive two weeks from onset of symptoms and has a high sensitivity [4]. These considerations, together with the rationale of limiting the number and duration of hospital access, call for the sequential combination of the two tests: the sensitive (and rapid) serology for screening and the specific (and slower) RT-PCR for diagnosis confirmation, as proposed in the algorithm presented above. However, conflicting results can occur, leading to complex decision-making. The case presented here may be interpreted either as a false-positive ELISA test or as an asymptomatic infection with positive IgM. Therefore, if the second hypothesis were true, questions arise. Are asymptomatic, IgM-positive patients at higher risk of post-operative complications? What types of precautions are needed when treating an IgM-positive patient?

Research on these and other questions related to cost-effectiveness of systematic screening should be encouraged to resume surgical activities optimally and safely in the COVID-19 era. Awareness of the time and effort required to generate the much-needed data, early experiences, and doubts could be beneficial to physicians facing similar situations.