Elective colorectal cancer surgery at the oncologic hub of Lombardy inside a pandemic COVID-19 area.

Of more than 209 000 individuals positive for severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) in Italy up until 2 May 2020, nearly 37% live in Lombardy, and of the 28 710 coronavirus disease 2019 (COVID‐19) related deaths, 49% occurred there. , To provide the best care, the Lombard healthcare system interrupted elective surgery in both community and regional hospitals, converting them into COVID‐19 referral centers. Therefore, patients affected by colorectal cancer on a waiting list for surgery in “hot” COVID‐19 areas suddenly were left without any chance to be treated. Thus, the National Cancer Institute of Milan (Lombardy, Italy), a comprehensive cancer center, was identified as an oncologic hub to provide surgical treatment for those patients. The mission was to dispose of that impressive waiting list as soon as possible, keeping our hospital “COVID‐free” in the safest setting at the same time. From 9 March to 24 April 2020 at the Colorectal Surgery Unit of the National Cancer Institute of Milan, a total of 54 patients affected by colorectal cancer were managed. Of these, 29 (53.7%) were cases from the institutional waiting list and 25 (46.3%) from the Lombardy oncologic hub list, coming mainly from the province of Bergamo. To manage patients from both hub and our institutional waiting list, prioritization was applied to each patient as following: red code (high priority, within 2 weeks) for complicated/symptomatic colorectal cancer (obstruction, perforation, bleeding); yellow code (medium priority, within 4 weeks) for cT2‐T4, N0‐N + colon or rectal cancer with indication for upfront surgery or rectal cancer awaiting more than 8 weeks after completion of neoadjuvant chemoradiation; green code (low priority, within 8 weeks) for cT1N0 colon cancer or positive margins after endoscopic polypectomy. The day before hospitalization, patients were called by phone to rule out symptoms compatible with COVID‐19 disease, such as fever and/or cough, and/or dyspnea. If no symptom was reported, patients were admitted to our Institute the following day to complete the dedicated triage, consisting in blood analyses, and a chest computed tomography (CT) scan. In case of lymphopenia, increased biomarkers of inflammation, and/or radiological ground‐glass opacities, patients were immediately tested for SARS‐CoV‐2 by nasopharyngeal swab and discharged. All other patients were admitted for hospitalization. Patients from the hub list were re‐evaluated by a restricted multidisciplinary colorectal cancer team. Laparoscopic approach was suspended to reduce the risk of aerosolization and the surgical time.

One (3.4%) patient on the institutional waiting list and seven patients (28.0%) on the hub list failed to pass the pre‐hospitalization triage because of chest CT scans positive for interstitial pneumonia (P = .019). Of the latter seven patients, four resulted negative for SARS‐CoV‐2. After a period of 15 days at home, they had regression or stability on CT scans and were admitted for hospitalization. The other three patients resulted positive for SARS‐CoV‐2 and were kept at home for a 28‐day quarantine: one suddenly developed pneumonia and died of COVID‐19; the other two patients remained asymptomatic, and after two negative swabs and evidence of regression of CT scan findings, were admitted for hospitalization. Clinical and pathological variables are reported in Table 1. To minimize or to defer surgery, in two cT1 and in one cT0 restaged after neoadjuvant chemoradiation for rectal cancer, endoscopic full‐thickness resection or trans‐anal local excision were carried out. A shorter surgical time in the hub patients depended on minimizing surgery, sometimes avoiding colorectal anastomosis, as suggested by persistent but milder chest‐CT findings observed in 28% of them. In fact, in three hub rectal cancer patients over 80 years of age who were not fit for chemoradiation, Hartmann's procedure was performed. Moreover, in three patients with obstructive rectal cancer, a loop colostomy was fashioned before neoadjuvant treatment. Only 6 out of our 12 colorectal team members performed surgery on hub patients, to limit the risk. Postoperative morbidity rates were similar in the two waiting list groups, despite a significantly higher age, white blood cells count, platelets, number of chest CT scan findings, longer time from diagnosis to surgery, lower pathological stages 0‐I (28.0 vs 48.2%, P = .382), worse ASA scores III‐IV (52.0 vs 27.5%, P = .095), and a red‐code priority level in the hub group. Particularly, postoperative interstitial pneumonia was observed in two hub (8.0%) and three institutional cases (10.3%). All these patients were temporarily moved to a dedicated observation ward and subjected to nasopharyngeal swabs, which were negative in all cases. The unexpectedly favorable clinical trend of the hub group deserves a closer look, since the Bergamo area was heavily affected by COVID‐19. Surgeons, nurses and all the other healthcare workers were daily assessed for symptoms (dry cough and/or fever ≥ 37.5°C) and admitted at work only if asymptomatic. During this period, no surgeon or nurse of our Colorectal Surgery Unit staff developed COVID‐19. As of 28 April 2020 all our staff members were tested for anti‐SARS‐CoV‐2 immunoglobulin G, and none resulted positive. Centralized management of colorectal cancer patients in an oncologic hub proved effective during the COVID‐19 outbreak.

Table 1

Clinical and pathological characteristics of patients

	Hub patients from COVID‐19 areas (n = 25)	Institutional patients (n = 29)	P value
Age at diagnosis, y	71.0 (±11.4)	63.0 (±16.0)	.045
Sex
Male	14 (56.0%)	12 (41.4%)	.413
Female	11 (44.0%)	17 (58.6%)
WBC count (x103/uL)	8.1 (±2.6)	6.7 (±4.0)	.018
Lymphocyte count (x103/uL)	1.6 (±0.4)	1.4 (±1.0)	.355
Platelet count (x103/uL)	285.4 (±62.5)	237.0 (±64.0)	.008
C‐reactive protein, mg/L	10.6 (±13.0)	16.1 (±62.9)	.67
Chest CT scan findings
Ground‐glass opacities	7 (28.0%)	1 (3.4%)	.016
Other non‐suspect opacities	3 (12.0%)	10 (34.5%)
No relevant findings	15 (60.0%)	18 (62.1%)
Preoperative CEA, ng/mL	5.8 (±10.6)	2.2 (±20.4)	.879
Preoperative CA 19.9, U/mL	14.0 (±4.4)	9.5 (±2.0)	.344
Priority level
Red: symptomatic cancer	15 (60.0%)	7 (24.1%)	.006
Yellow: stage II‐III cancer	5 (20.0%)	18 (62.1%)
Green: early cancer	5 (20.0%)	4 (13.8%)
ASA score
II	12 (48.0%)	21 (72.4%)	.183
III	11 (44.0%)	7 (24.1%)
IV	2 (8.0%)	1 (3.4%)
Cancer localization
Right colon	4 (16.0%)	4 (13.8%)	.817
Transverse colon	4 (16.0%)	2 (6.9%)
Left/sigmoid colon	2 (8.0%)	4 (13.8%)
Rectum	13 (52.0%)	16 (55.2%)
Others	2 (8.0%)	3 (10.3%)
Clinical T stage
cT0‐1	4 (16.0%)	4 (13.8%)	.488
cT2	3 (12.0%)	8 (27.6%)
cT3	12 (48.0%)	13 (44.8%)
cT4	6 (24.0%)	4 (13.8%)
Clinical N stage
cN0	12 (48.0%)	17 (58.6%)	.585
cN1‐2	13 (52.0%)	12 (41.4%)
Clinical M stage
M0	22 (88.0%)	29 (100.0%)	.093
M+	3 (12.0%)	0 (0.0%)
Neoadjuvant treatment
No	22 (88.0%)	21 (72.4%)	.191
Yes	3 (12.0%)	8 (27.6%)
Surgery
Endoscopic full‐thickness resection	2 (8.0%)	1 (3.4%)	.592
Right colectomy	3 (12.0%)	4 (13.8%)
Transverse colectomy	2 (8.0%)	2 (6.9%)
Left/sigmoid colectomy	2 (8.0%)	4 (13.8%)
Anterior resection	10 (40.0%)	16 (55.2%)
Others/colostomy fashioning	5 (20.0%)	2 (6.9%)
Death before surgery	1 (4.0%)	0 (0.0%)
Ostomy
Yes	8 (32.0%)	9 (31.0%)	1.000
No	17 (68.0%)	20 (69.0%)
Surgery time, min	141.4 (±61.8)	191.0 (±71.0)	.009
Postoperative stay, d	6.9 (±3.8)	8.5 (±5.0)	.197
Complications
Postoperative ileus	1 (4.0%)	0 (0.0%)	1.000
Anastomotic leakage	2 (8.0%)	5 (17.2%)
Bleeding	0 (0.0%)	1 (3.4%)
Ureteral leakage	1 (4.0%)	0 (0.0%)
No complications	20 (80.0%)	23 (79.4%)
No surgery	1 (4.0%)	0 (0.0%)
(y)pT stage
pT0‐1	6 (24.0%)	5 (17.2%)	.132
pT2	1 (4.0%)	9 (31.0%)
pT3	10 (40.0%)	10 (34.5%)
pT4	4 (16.0%)	4 (13.8%)
Not applicable	4 (16.0%)	1 (3.4%)
pN stage
pN0	13 (52.0%)	19 (65.5%)	.555
pN1‐2	8 (32.0%)	8 (27.6%)
Not applicable	4 (16.0%)	2 (6.9%)
Grading
G1	6 (24.0%)	4 (13.8%)	.433
G2	12 (48.0%)	20 (69.0%)
G3	3 (12.0%)	3 (10.3%)
Not applicable	4 (16.0%)	2 (6.9%)
Margins status
R0	19 (76.0%)	27 (93.1%)	.426
R1	1 (4.0%)	0 (0.0%)
Not applicable	5 (20.0%)	2 (6.9%)

Abbreviations: CA, carbohydrate antigen; CEA, carcinoembryonic antigen; COVID‐19, coronavirus disease 2019, CT, computed tomography.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.