Introduction

The World Health Organization (WHO) has classified the ongoing global pandemic of viral pneumonia that has been caused by the rapid spread of the novel coronavirus 2019-nCov (COVID-19) – designated as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) as a public health emergency¹,². Mortality rate for patients with severe or critical new coronavirus pneumonia increased with age and with the presence of more comorbid conditions, such as diabetes, cardiovascular disease, and cerebrovascular illness²,³. Notably, the systemic immunosuppressive state brought on by cancer puts patients at higher risk for serious infections and in a worse prognosis.

For colon cancer patients with coronavirus pneumonia, a number of conservative therapy approaches are currently advised, although surgery for individuals with surgical reasons is typically postponed³. For the majority of patients with SARS-CoV-2 infection, it is advised to stop cancer treatment if the infection worsens. Due to this, cancer patients had a comparatively high chance of tumor progression, which raised questions about the efficacy of tumor therapy⁴. Safe implementation of colorectal cancer surgery during the pandemic is challenging due to safety issues of the operating rooms and surgery team members, the restrictions of a limited number of hospital services and health staff, and the concerns of post-operative COVID-19-associated complications and mortality⁵. Therefore, there is a need to minimize the risk of COVID-19- positive patients for elective surgery and to introduce prioritization and additional safety measures in clinical practice.

Cancer initiation, progression, and metastasis are not only dependent on the type of cancer, but the systemic inflammation and nutritional status play important roles. The prognosis screening in colon cancer is crucial due to high morbidity and mortality rates. There are several indexes and classification of colon cancer patients. One of these indexes, hemoglobin, albumin, lymphocyte, and platelet (HALP) score, is an inflammation index that has been reported to be beneficial in showing the prognosis of stomach, colorectal, kidney, bladder, and prostate cancers⁶. Other indexes are the pre-operative Frailty index (FI) and Clavien Dindo Classification (CDC), which are valuable indicators of mortality and morbidity, respectively. Another index is the Charlson comorbidity index (CCI), which is the most used comorbidity index⁷,⁸. CCI categorizes the comorbidities of patients based on the International Classification of Diseases diagnosis codes found in administrative data, such as hospital abstracts data. Each comorbidity category has an associated weight¹-⁶, based on the adjusted risk of mortality or resource use, and the sum of all the weights results in a single comorbidity score for a patient (CCI; < 4 and ≥ 4). CCI has been adapted and verified as applicable and valid for predicting the outcome and risk of death from the colon cancer⁹. Regarding various studies based on all these indexes in colon cancer patients during the COVID period are limited.

Considering the prognostic indices, we aimed to compare the findings before and after COVID-19 to predict the pre- and post-operative prognostics of the colon cancer patients based on the elective and emergency colectomies.

Method

Results

Demographic and baseline features of patients are presented in table 1. There were 188 patients who underwent surgery before the pandemic and 62 patients during pandemic due to the diagnosis of colon cancer. In pre-pandemic period, 87 patients (46.3%) were operated in emergency conditions and 101 (53.7%) were operated in elective conditions. However, during pandemic, 30 patients were emergency cases, and 32 were elective cases (48.4% vs. 51.6%, respectively, p = 0.820).

Before the pandemic, 31 patients in 2017, 62 patients in 2018, 81 patients in 2019, and 14 in March 2020, a total of 188 patients were operated. During the pandemic, 32 patients were operated after March 2020, and 30 patients were operated until 19 August 2021. Most of colon cancer patients were male (57.6%), 65 were emergency cases, and 119 were elective cases. Mean age of all patients was 63.02 ± 12.48 years. No statistical significance was found for the distribution of age and gender for total (p > 0.05) (Table 1).

Majority of pre-COVID patients were at ASA III (44.1%), and of post-COVID patients were at ASA II (51.6%). Since the number of emergency surgeries in the pre-pandemic period was more than those in the post-pandemic period, the ASA score of pre-COVID patients was found as higher and statistically significant compared to that of post-COVID patients (p = 0.018). However, comparison of ASA scores on the basis of emergency and elective condition, there was no significant difference between the scores of pre-pandemic and pandemic period (p = 0.057 vs. 0.304, respectively) (Table 1).

BMI of all colon cancer patients and that of elective cases significantly differed between pre- and post-pandemic periods (p = 0.025 vs. 0.049, respectively), but BMI of emergency cases did not differ significantly (p = 0.369). No statistical significance was found in major/minor complications for all patients and different conditions (p > 0.05). Most emergency cases in pre- COVID period had FI score 0, while most during the pandemic had score over 2 (44.8% vs. 43.3%, respectively; p = 0.360). Most of the elective cases in pre-COVID period had FI score over 2, while most during pandemic had score 0 (40.6% vs. 43.8%, respectively, p = 0.393). Most of all cancer patients and of emergency cases were classified at CDC II (43.2% and 41.4%, respectively), and there was a significant difference in the distribution of CDC classification of pre- and post-COVID patients (p = 0.014 and 0.006, respectively). However, CDC classification of elective cases did not significantly differ (p = 0.329). No significance was found in the mean CCI values of all patients independent of the operation conditions (p > 0.05) (Table 1).

The laboratory findings of all patients were presented in table 2. The mean albumin, CRP, WBC, lymphocyte, neutrophil, thrombocyte, hemoglobin, MPV, and HALP score, which showed no significant difference between periods for all patients and also for emergency and elective cases (p > 0.05), except that there was a significance in WBC values of the emergency cases (p = 0.040).

Considering the clinicopathology of patients (Table 3), the tumor was most commonly located in the left colon in pre-COVID period among all patients (59.0%), the emergency cases (62.1%), and elective cases (56.4%), which did not differ compared to post-COVID patients (p > 0.05). However, the most frequently performed operation was right hemicolectomy in patients who could be operated in both periods (41.0% vs. 46.8%, respectively, p = 0.743). Type of procedure did not differ significantly among all patients and based on the operation conditions (p > 0.05). Most patients did not need a stoma (75.0% vs 86.9%, respectively, p = 0.107). Rate of ileus formation did not differ significantly among all patients and based on the operation conditions (p > 0.05). 16.1% of the emergency cases before the pandemic had ascites, while none of those in post- pandemic period had ascites (p = 0.044). No difference was observed for the rate of ascites formation among all patients and elective cases (p > 0.05) (Table 3).

Mean tumor diameter and distribution of lymphatic invasion and adjacent organ invasion did not differ among all patients and patients operated under elective conditions (p > 0.05); however, the rate of perineural invasion among patients operated under elective conditions was significantly higher during the pandemic compared to pre-pandemic (40.6% vs. 19.8%, p = 0.032). Grade 2 tumors and T4 tumors predominated in both periods (72.3% vs. 80.6% and 48.4% vs. 56.5%, respectively) without any significant difference among groups (p > 0.05). moreover, the most common stage of tumor among all patients was Stage 3, observed in both periods (42.6% vs. 48.4%, respectively) without any significant difference (p = 0.284). N1 tumors (38.8%) were in the majority in pre-pandemic period, while number of N0 tumors (41.9%) was higher during COVID, without any significant difference (p = 0.525). The distribution of grade, TN stage, and overall stages of the emergency cases and elective cases did not differ significantly between pre-pandemic and pandemic period (p > 0.05). However, the rate of solid organ metastasis (M) of all patients was significantly higher in pre-pandemic period compared to that during pandemic (19.7% vs. 11.3%, p = 0.046). Emergency cases was also significantly higher in pre-pandemic period (35.6% vs. 13.3%, p = 0.039), but the rate of LN metastasis did not differ among periods (p > 0.05) (Table 3).

The mortality rate was 30.4% (n = 76) among all 250 patients; 36 of 117 emergency cases (30.8%) and 40 of 133 elective cases (30.1%) died due to the pulmonary issues, anastomosis leakage, myocardial infarcts, multiple organ failure, etc. The mortality rate was 37.2% before the pandemic, but 9.7% during the pandemic, which was significantly lower than the pre-pandemic rate (n = 70 vs. n = 6, respectively, p < 0.001). The mean survival duration was 20.39 ± 14.81 months for all patients; 24.39 ± 14.74 months before the pandemic and 8.24 ± 5.51 months during the pandemic, which was significantly lower than the mean of pre-pandemic survival duration (p < 0.001). According to Kaplan-Meier estimation results, there was no significant association between pandemic period as an independent prognostic factor for survival rates of the colon cancer patients (HR=0.965, 95% CI: 0.417-2.568, p = 0.94) (Fig. 1).

Figure 1 Cancer-specific survival curve (Kaplan-Meier curve) for patients with colon cancer during COVID-19 pandemic period. 

The Multivariate Cox regression analysis of clinicopathological factors showed that an increase in CDC (OR: 0.123; 95% CI: 0.331-1.393; p = 0.007) and in CCI (OR: 0.067; 95% CI: 0.147-1.158; p = 0.028) primarily affected the mortality rate during COVID-19 (Table 4). However, other clinicopathological factors, including age, BMI, HALP, grade, and stage of colon cancer, location of tumor, FI, ASA score, type of procedure, and surgery did not affect the mortality rate significantly during pandemic (Table 4).

Discussion

During COVID-19, both elective and emergency surgeries have been canceled or postponed in cancer patients²,¹³,¹⁴. The current study included patients with colon cancer that underwent elective or emergency surgeries before and during the pandemic. We found that FI and HALP scores did not differ among periods, while CDC grade was significantly higher among all patients and especially in elective cases before COVID-19. Frequency of ascites and tumors at M stage was significantly higher among emergency cases before COVID-19, while the frequency of perineural invasion was significantly higher during COVID-19 among elective cases. CDC and CCI scores primarily affected the mortality rate during COVID-19.

During COVID-19, the risks and benefits between the protection of patients and health care providers and the adverse outcomes resulting from delayed surgery need to be balanced. Particularly consistent criteria for "emergent" colorectal surgery should be established that could help to simplify surgeons’ decisions to operate in their practices¹⁵,¹⁶. Consistency with the reports, there has been an increase in the number of urgent surgeries at hospitals where elective activity was reduced¹⁷,¹⁸, as in our practices.

Age and gender, as patient-level factors, along with overall health status and comorbidities, are included in clinical risk scores associated with worse outcomes of SARS-CoV-2 infection¹⁹. The evaluation of Kuzuu et al. in pre–COVID-19 and during pandemic reported without significant differences between two periods in terms of the proportions of gender and age according to cancer²⁰. In the current study, most of colon cancer patients in pre-COVID and post-COVID period were male, and the mean age in post-COVID period was slightly lower without a significant difference between periods and ineffective for mortality rates during pandemic.

Studies indicated that the numbers of elective colorectal surgeries during the pandemic did not differ in age, gender, and BMI between the number of elective surgeries in two years before pandemic²¹,²². Inconsistently, BMI of 250 colon cancer patients and that of 133 elective cases significantly differed between pre- and post-pandemic periods, but BMI of 117 emergency cases did not differ significantly, suggesting that BMI did not compromise a bias for surgeries for emergency cases but may be important point of view in the elective surgeries. Similar to the literature, we eventually reported that BMI was not an effective factor for the mortality during COVID-19 pandemic.

The diagnostic, prognostic factors and indexes in colon cancer patients have become more apparent during pandemic. Effects of COVID-19 on some outcomes for some cancer types in 2019 and 2020 were investigated and reported that more patients presented with ASA level 3 in the COVID-19 year versus 2019²³. Similarly, ASA score 3 was mostly reported in pre-COVID period, while ASA score 2 was seen promptly in post-COVID period. However, the evaluation of emergency and elective surgeries regarding ASA scores between two periods showed no any significant correlation in the mortality rate during pandemic.

Studies have indicated higher mortality and complications suffered by patients undergoing surgery during pandemic, particularly those with SARS-CoV-2 infection²⁴. Moreover, it was reported that some patients avoided visiting the hospital unless it was an emergency because they feared contracting COVID-19²⁵. The need for this group was motivated by widespread concern among surgical patients about the unanticipated consequences of COVID-19, which had an impact on morbidity and mortality rates²⁶. It was reported that the cancer, cardiac disease, chronic immunosuppression, chronic anticoagulation, reoperation, intraoperative and post-operative blood transfusion, or the duration of surgery were significantly associated with complications. In addition, a significant correlation between complications and other relative risk variables was found in the post-operative infection²⁷. In another study, colorectal cancer patients showed no significant differences in comorbidities, CCI, and mortality rate determined in pre-COVID vs. during COVID²⁸. According to our study, the minor complications, including wound infection, ileus, wound dehiscence, urinary tract infection, and major complications including anastomotic leakage, intra-abdominal abscess, sepsis, pneumonia, pulmonary embolism, acute renal failure, myocardial infarction, cardiac arrest, cerebrovascular disease, the need for intubation and reoperation were less likely higher in pre-COVID period without a statistical significance. During the pandemic, there were also no differences in post-operative recovery and incidence of post- operative complications. Surgical duration, amount of intraoperative bleeding, and length of post- surgical hospital stay did not change substantially. This could be attributed to a combination of enhanced recovery after surgery (ERAS) guidelines and laparoscopic minimally invasive surgery, which has a slight impact on post-operative nutritional status and shortens the length of post- operative hospital stay, especially in older patients with colorectal cancer, as in literature²⁹,³⁰. Consistently, CDC Grade 2 was promptly seen in both periods in patients with morbidity. These differences were eventually apparent for all colon cancer patients and specifically for emergency cases. Interestingly, CDC was a significant factor affecting the mortality during pandemic, suggesting that CDC is as a promising feature to be predicting the prognosis of colon cancer patients.

Colorectal cancer is largely diagnosed at old age, when comorbidities and frailty are common and might be important prognostic factors of this cancer, which are strong prognostic factors of survival in colorectal patients apart from commonly considered sociodemographic and tumor characteristics. In a meta-analysis for the impact of comorbidity and frailty on prognosis in colorectal cancer patients, compared to colorectal cancer patients without comorbidity, those with mild/moderate and severe comorbidity were found to have a higher risk of 30-day, overall, and CRC-specific mortality. In addition, frail CRC patients showed higher overall mortality than non-frail patients³¹. In some recent studies, the relationship between the mFI-5, malnutrition, body composition, systemic inflammation, and short-term clinical outcomes in patients undergoing surgery for colorectal cancer was investigated and found that the male gender elevated inflammation, and mFI-5 score remained significantly associated with the incidence of post- operative complications. They reported that mFI-5 frailty score was significantly associated with age, systemic inflammation, and post-operative outcomes in patients undergoing potentially curative resections for colorectal cancer¹². We also did not find a significant difference in both CCI and frailty indexes between pre- and post-COVID periods, but CCI was found to be correlated with increased mortality during pandemic.

Current coronavirus evolution models and prognosis scores have used in evaluation, including HALP scoring and management of severe cases to predict the prognosis of disease³²,³³. In a very recent review, the prognostic role of HALP score on survival outcomes in patients with solid tumors was investigated. Low HALP score was associated with decreased overall survival, cancer-specific survival, and progression/disease/recurrence-free survival. The prognostic value of HALP on overall survival was observed across various tumor types and tumor stages. They concluded that a low pretreatment HALP score is a reliable and negative prognostic biomarker for survival outcomes in patients with cancer³⁴. Therefore, we investigated HALP score between pre- and post- COVID periods unlikely with a significant difference for emergency and elective cases.

Various cancer screening programs for colorectal cancer have been suspended in many countries after the onset of the COVID-19, such as the completion of colonoscopies with declines³⁵. Therefore, we compared the survival duration during the pandemic, which was significantly lower than the mean of pre-pandemic survival duration. However, the results of survival analysis showed a less likely an association between pandemic period as an independent prognostic factor for survival rates of colon cancer patients. There are controversy findings whether an appropriate delay in the surgery date among patients with Stage I to III colorectal cancer has an influence on survival time. A study concluded that an appropriate delay of surgical date did not affect survival³⁶. In fact, any delay in treatment may lead to tumor progression and an increased risk of transformation from a resectable to an unresectable tumor. Some researches predicted that delays in diagnosis brought about by the pandemic would shorten the long-term survival time of colorectal cancer patients, potentially increase in mortality within 5 years of diagnosis³⁰. Including the stages at diagnosis among patients with gastrointestinal cancer during COVID-19 was evaluated with a significant decrease in numbers of patients with stage 0 or I colorectal cancer²⁰. Considering post-operative clinicopathology of our cohort, even though the distributions of patients according to the grade, TN stage and overall stage of cancer were comparable before and during pandemic, the proportion of patients with perineural invasion (a feature of adverse prognosis) among elective cases significantly increased during pandemic, suggesting that clinical diagnosis and treatment of colorectal cancer may proceed following the epidemic prevention³⁶. Clinically, our patients had more likely slight increases in tumoral extension and staging.

There are some limitations based on the data; (1) nested regression models to generate some imputations for missing values < 30 % of laboratory results, (2) retrospective analysis including measurement and recall biases, (3) without stepwise or any other machine learning models. However, this is the first and largest study to identify the factors to predict mortality and post-operative complications in early settings for elective and emergency colon cancer surgery. This makes it a distinctively proficient tool that can guide surgeons primarily to distinguish the patients to prompt surgical interventions regarding physiological and pre-operative assessments.

Conclusions

We investigated the effective factors for mortality of patients who underwent colonic resection for cancer during pandemic. The operative management of comorbidities, CDC for the complications, and CCI for comorbidities might be a helpful clinical tool to better identify the post-operative mortality regarding pandemic, priorly. In addition, the prediction of post-operative outcomes when accounting for patient comorbidities and patient acuity might add value to current challenges during the pandemic to improve the quality of care and decrease the mortality. In the future, incorporation of systemic screening tools in clinics may help to improve the prognosis of colon cancer patients.