Has COVID-19 Modified the Weight of Known Systemic Inflammation Indexes and the New Ones (MCVL and IIC) in the Assessment as Predictive Factors of Complications and Mortality in Acute Pancreatitis?

We aimed at evaluating the prognostic capacity of the inflammatory indices derived from routine complete blood cell counts in two groups of patients with acute pancreatitis from two different time periods, before and during the COVID-19 pandemic, when a high incidence of complications with surgical risk and mortality was found. Two new markers were introduced: the mean corpuscular volume to lymphocyte ratio (MCVL) and the cumulative inflammatory index (IIC), which were calculated at a baseline in the two groups of patients. Of the already established markers, none of them managed to effectively predict the complications with surgical risk and mortality, with a decrease of less than 50% in specificity in the peri-COVID group. The MCVL had the best prediction of complications with surgical risk in both the pre-COVID and peri-COVID groups, validated it as an independent factor by multivariate analysis. The IIC had the best prediction of mortality in both periods and was proven to be an independent factor by multivariate analysis. As the IIC predicted death best, we tested the occurrence of death and found that patients with PA who had an IIC > 12.12 presented a risk of death 4.08 times higher in the pre-COVID group and 3.33 times higher in the peri-COVID group. The new MCVL and IIC independent markers had a superior sensitivity and specificity in predicting surgical risk complications and, respectively, mortality in the group of patients with acute pancreatitis during the COVID-19 pandemic, which makes them widely applicable in populations with modified immune and inflammatory status. Conclusions: In patients with acute pancreatitis, MCVL has a significant predictive value regarding complications with surgical risk (abscess, necrosis, and pseudocyst), and the IIC has a significant predictive value for mortality.


raiova
200349
raiovaRomania

Eugen Florin Georgescu 0000-0003-4266-4131
General Surgery Department
University of Medicine and Pharmacy of Craiova
200349CraiovaRomania

Costin Teodor Streba 
Department of Pneumology
University of Pharmacy and Medicine Craiova
200349CraiovaRomania

Razvan Mercut razvanmercut@gmail.com 
Department of Plastic and Reconstructive Surgery
University of Medicine and Pharmacy of Craiova
200349CraiovaRomania

Elena Irina Caluianu 
General Surgery Department
University of Medicine and Pharmacy of Craiova
200349CraiovaRomania

Emil Tiberius Trasca 
General Surgery Department
University of Medicine and Pharmacy of Craiova
200349CraiovaRomania

Dumitru Radulescu 
General Surgery Department
University of Medicine and Pharmacy of Craiova
200349CraiovaRomania

Has COVID-19 Modified the Weight of Known Systemic Inflammation Indexes and the New Ones (MCVL and IIC) in the Assessment as Predictive Factors of Complications and Morta ity in Acute Pancreatitis?
10 December 2022D6DD7D3AD00A14C6D74666907DD391FF10.3390/diagnostics12123118Received: 19 October 2022 Accepted: 8 December 2022Radulescu, P.M.Davitoiu, D.V.Baleanu, V.D.Padureanu, V.Ramboiu, D.S.Surlin, M.V.Bratiloveanu, T.C.Georgescu, E.F.Streba, C.T.Mercut, R.et al. Has acute pancreatitisIICcumulative inflammatory indexMCVLaverage crepuscular volume−lymphocyte ratioabscessnecrosispseudocystNLRneutrophil-lymphocyte ratio
We aimed at evaluating the prognostic capacity of the inflammatory indices derived from routine complete blood cell counts in two groups of patients with acute pancreatitis from two different time periods, before and during the COVID-19 pandemic, when a high incidence of complications with surgical risk and mortality was found.Two new markers were introduced: the mean corpuscular volume to lymphocyte ratio (MCVL) and the cumulative inflammatory index (IIC), which were calculated at a baseline in the two groups of patients.Of he already established markers, none of them managed to effectively predict the complications with surgical risk and mortality, with a decrease of less than 50% in specificity in the peri-COVID group.The MCVL had the best prediction of complications with surgical risk in both the pre-COV D and peri-COVID groups, validated it as an independent factor by multivariate analysis.The IIC had the best prediction of mortality in both periods and was proven to be an independent factor by multivariate analysis.As the IIC predicted death best, we tested the occurrence of death and found that patients with PA who had an IIC > 12.12 presented a risk of death 4.08 times higher in the pre-COVID group and 3.33 times higher in the peri-COVID group.The new MCVL and IIC independent markers had a superior sensitivity and specificity in predicting surgical risk complications and, respectively, mortality in the group of patients with acute pancreatitis during the COVID-19 pandemic, which makes them widely applicable in populations with modified immune and inflammatory status.Con lusions: In patients with acute pancreatitis, MCVL has a significant predictive value regarding complications with surgical risk (abscess, necrosis, and pseudocyst), and the IIC has a significant predictive value for mortality.

Introduction

Globally, the incidence of acute pancreatitis s continuously increasing; in many countries, it is a significant health burden, affecting 34 people out of 100,000 per year [1].The incidence of acute pancreatitis varies in Europe between 4.6 per 100,000 inhabitants in Albania and 100 per 100,000 inhabitants in Poland [2].Depending on the biliary and alcoholic etiology of acute pancreatitis, the studies showed ratios of 6:1 in the Mediterranean countries of southern Europe [3], while the lowest ratio between gallstones and alcohol was less than 0.5:1 and was found mainly in Eastern European countries, including Romania [4].The diagnosis of drug-induced acute pancreatitis is difficult to establish, with an incidence of less than 2%, and it is underreported due to the technical difficulty of diagnosis [5].Most European studies carried out in the last five decades showed an increase in incidence by approximately 3-4% per year [6].

Acute pancreatitis is defined as an acute inflammatory condition that puts life at risk, ranging from a simple edema to necrosis, through the intraglandular activation of pancreatic enzymes.According to the Atlanta classification system, acute pancreatitis is divided into mild, moderate, and severe, depending on the severity [7].In most cases, 80-90% have mild and moderate acute pancreatitis.On the other hand, 10-20% of cases have severe disease, w ere mortality is 30-50% [1].Its evolution comprises two stages, the initial one in the first week, which is accompanied by organ failure, and the late one after the first week, which is accompanied by local complications, such as infected necrosis, abscess, or pseudocyst formation [8].Improvements in early and accurate diagnosis, as well as the care of people with severe acute pancreatitis, led to a decrease in the overall mortality to 0.8%; still, this aspect, as well as the long-term effects, remains significant [9].

Coronavirus disease-19 (COVID-19) first occurred in Wuhan, China; it was declared a pandemic in March 2020 and subsequently became a major global health threat.Since 2021, over 100 million people from 210 countries have been confirmed to have been infected with the COVID-19 virus [10].

In Romania, the first case of infection with the new coronavirus was confirmed on 26 February 2020, and the first 3 deaths were recorded on 22 March 2020, reaching a maximum of 591 deaths per day on 2 November 2021.The virus is expe ted to remain active until 2024 even after adequate control measures, especially with the emergence of many mutants called variants of concern, such as B.1.1.7 (alpha), B.1.617.2 (delta), and B.1.1.529(omicron) [11].Liu et al. studied the damage to the pancreas by the SARS-CoV-2 infection and the association with ACE-2 receptors, which are slightly more expressed in the pancreas than in the lungs and may lead to the direct damage to the pancreas that was reported in 2% as non-severe, compared to 17% of the patients with severe COVID-19 [12].

To predict the prognosis and severity of PA, serum markers such as procalcitonin, C-reactive protein (CRP), N

trophil CD64 Index,
interleukin-6, and interleukin-8 were used, but they are expensive, barely available, and cannot accurately predict the prognosis and severity of PA [13,14].

For the past two decades, complete blood cell counts have routinely been performed, and they are relatively inexpensive and easy to perform.Neutrophils and monocytes are components of the innate immune system; lymphocytes are indicators of the adaptive immune response, and platelets, in addition to hemostasis, coagulation, and angiogenesis, are involved in the inflammatory reaction and innate immunity [15].The hematological indices derived from the leukocyte count are: the neutrophil/lymphocyte ratio (NLR) based on the coexistence of leukocytosis and lymphopenia in the initial in

the derived n
utrophil/lymphocyte ratio (dNLR); the monocyte/lymphocyte ratio (MLR), reflecting the body's immune status, with a decrease indicating an immune dysfunction of the host; the platelet/lymphocyte ratio (PLR), which is an inflammatory marker of immune-mediated, metabolic, and prothrombotic diseases; the aggregate index of systemic inflammation (AISI), which is studied in lung diseases [17] and COVID-19 [18]; and the systemic inflammatory response index (SIRI) and th systemic inflammatory index (IIS), which are new inflammatory markers that are increasingly studied as biomarkers in several diseases, such as neoplastic [19], inflammatory, and infectious diseases such as acute pancreatitis [20] and COVID-19 [21].

Currently, there are few studies regarding their importance when taken separately in predicting severity and outcomes, and they were not analyzed together in predicting complications with surgical risk such as abscess, necrosis, pseudocyst, and mortality, either in the period before the pandemic or during the COVID-19 pandemic.Considering that, lately, there has been an emphasis on the identification of new biomarkers w th improved predictive performances, we limited ourselves to the simple analysis of the peripheral blood upon admission to the hospital in the first 24 h, without performing any additional biological investigations.Following the observation of increased values of RDW and MCV upon hospitalization in the patients who died during hospitalization, we developed two new markers: the MCVL (the ratio between the average corpuscular volume and lymphocytes) and the IIC (cumulative inflammatory index).They reflect the numerical changes that occur at the level of red and white blood cells following an inflammatory process.

The aim of the study is to assess the values of NLR, PLR, MLR, dNLR, AISI, SIRI, SII, and the newly introduced MCVL and IIC as independent predictive factors in surgical risk complications and mortality in patients with AP and to test their accuracy in two time periods, before COVID-19 (pre-COVID) and during the COVID-19 pandemic (peri-COVID), when cell-mediated immunity is affected [22] and can decrease their predictability.The establishment of reliable predictive markers may help to prompt the initiation of medical management and thus obtain better outcomes for the patients with AP in terms of surgical risk complications and mortality.(Table 1) Table 1.Research questions.


Research Questions

Did the number of patients diagnosed with AP decrease during the COVID-19 pandemic, and what would the explanation for this be?Did the mean values of the already known inflammatory markers NLR, PLR, MLR, dNLR, AISI, SIRI, and SII and the newly introduced MCVL and IIC change during the COVID-19 period?Were there any differences in terms of the number of complications with surgical risk and their degree of operability during the COVID-19 period?Among the already known inflammatory markers NLR, PLR, MLR, dNLR, AISI, SIRI, and SII and the newly introduced MCVL and IIC, which of them can effectively predict the complications with the surgical risk (abscess, necrosis, and pseudocyst) and mortality in the pre-COVID and peri-COVID periods?


Materials and Methods


Study Design

We retrospectively studied the consecutive PA cases from two academic medical centers, the Emergency County Clinical Hospital of Craiova and the Military Emergency Clinical "Dr.Stefan Odobleja" Hospital of Craiova, after obtaining the approval of the Ethics Commission of each unit.We used the hospital automation systems to collect patient data by ICD-10 diagnostic code (K85), including for the study cases that had enough data to be validated in their entirety.

After applying the inclusion criteria, the data of 433 patients with acute pancreatitis admitted to the surgery service were analyzed; they were then divided into two groups, one before the pandemic, including patients admitted from 1 January 2018 to 26 February 2020 (pre-COVID), when the first case of COVID-19 was officially declared in Romania, and a second group that included patients from 27 February 2020 to 30 April 2022 (peri-COVID) (Figure 1).The data collected and considered were the following: demographic data (sex and age), the number of hours since the onset of symptoms, and the degree of severity obtained by correlating clinical and imaging data [23].

admitted to the surgery service were analyzed; they were then divided into two groups, one before the pandemic, including patients admitted from 1 January 2018 to 26 February 2020 (pre-COVID), when the first case of COVID-19 was officially declared in Romania, and a second group that included patients from 27 February 2020 to 30 April 2022 (peri-COVID) (Figure 1).The data collected and considered were the following: demographic data (sex and age), the number of hours since the onset of symptoms, and the degree of severity obtained by correlating clinical and imaging data [23].The result of the first blood test in the first 24 h after presentation was also taken into account and included: the number of white blood cells (WBC), the numbe

of neutrophils (NEU
, the number of lymphocytes (LYM), the number of monocytes (MON), the platelet count (PLT), the red cell distribution width (RDW), and the mean corpuscular volume (MCV).Having obtained the laboratory tests available after the first blood collection, we collected the data obtained from the blood cell count in the peripheral blood at admission, and we calculated the inflammation indices derived from t

ocytes ×
000)).

The etiological factors were obtained by examining the data recorded from the anamnesis (alcohol intake), the family history (autoimmune diseases), and the routine imaging examinations.Patients without an etiological factor were registered as "unknown".Data such as number of hospitalization days, complications with surgical risk, and mortality were recorded.We included patients ho had undergone procedures such as surgery for complications and medical therapy for AP treatment.The diagnosis of COVID-19 was established by real-time reverse transcription polymerase chain reaction (RT-PCR) nasopharyngeal swab testing.Patients who had a negative result in two tests were considered negativ according to the recommendation of the Romanian Ministry of Health.Depending on the results, the patients in the peri-COVID group were divided int

two groups, the COVID-19 (n = 28
and the non-COVID-19 (n = 168) group, with data comparison between these two groups.


Inclusion Criteria

The patients included in the study were those with the diagnosis of acute pancreatitis, confirmed biologically and by imaging according to the revised Atlanta Criteria [24], from 1 January 2018 to 30 May 2022, including the first blood analysis performed before starting medical treatment and those who had completed chemotherapy, radiotherapy, or combined treatment for various neoplastic conditions.


Exclusion Criteria

Patients excluded from the study were those with incomplete data at admission and a chronic treatment of oral corticosteroids and immunosuppressant drugs in the previous three months; those with incomplete oncological treatment; those with autoimmune diseases; and pregnant women.


Statistical Analysis

Data analysis was performed using SPSS version 25.0 (released by IBM Corp in 2017), with the establishment of descriptive statistics of the studied population.

The data were first tested for normality and homogeneity of variation.The values that were normally distributed were expressed as ±SD, and categorical values were expressed as percentages.The independent t-test was performed to determine the effect of the inflammatory indices on complications or mortality.

The optimal value was established following the analysis of the characteristic ROC curve, determining a maximum normal limit in order to maximize the specificity and sensitivity of each index.Limits were set by maximizing the sum of the sensitivity and the specificity.

The results were expressed as mean value ± standard deviation for continuous quantitative variables.The quantitative variables were expressed in the form of proportions.The univariate statistical analysis was performed usi g hypothesis confirmation tests: the chi-squared test for qualitative variables and the Student t test for comparing quantitative variables with the homogeneity of variations in the Levene test, whose results are less affected by unequal group sizes [25].To adjust for potential confounding effects, multivariate logistic regression analyses were performed, with 95% confidence intervals (CIs).We considered the results to be statistically significant if the p values of the two groups were <0.05, with a confidence interval of 95%.


Results

We retrospectively analyzed a total of 538 patients presented to the emergency room and diagnosed with acute pancreatitis between 1 January 2018 and 30 May 2022.Among these patients, we included in the study a total of 433 patients who met the inclusion criteria and on whom there were enough data from the laboratory analyses performed in the first 24 h after admission o the hospital.

The patients were divided into two groups; the first group was called pre-COVID and included patients admitted between 1 January 2018 and 3 March 2020 (n = 237), and the other group we called peri-COVID; it included patients admitted during the COVID-19 pandemic between 3 March 2020 and 5 May 2022 (n = 196).

The patients in the peri-COVID group were also divided into two groups: positive for COVID-19 (COVID-19) and negative for COVID-19 (non-COVID-19).


Characteristics of the Patients

Among the patients in the pre-COVID group, 56.5% were men and had a mean age of 57.57± 13.43 (n = 134), while the mean age of the female patients was 58.78 ± 19.89 years (n = 103) (Table 2).Regarding the etiology of pancreatitis, the most common was gallstones 82.2% (n = 195), followed by alcohol intake 12.6% (n = 30), and unknown etiology 4.64% (n = 11).According to the Atlanta classification, 67.5% (n = 81) had a mild form, 46.7% (n = 86) had a moderate form, and 54.3% (n = 70) had a severe form of PA.Regarding the complications with surgical risk, the was total was 11.8% (n = 28), represented by abscess 3.8% (n = 9), necrosis 3.8% (n = 9), and pancreatic pseudocyst 4.2% (n = 10); in 6.8% (n = 16) of the cases, there were complications (6.8%) requiring surgical intervention.The mean duration of hospitalization was 14 ± 16.55 days, and discharge occurred in 10.1% of the patients (n = 24).In the peri-COVID group, 46.4% were men and had a mean age of 62.19 ± 17.21 (n = 91), while the mean age of the female patients was 65.55 ± 15.45 years old (n = 105).Regarding the etiology of pancreatitis, the most common was gallst

o was higher in the pre-COVID group (1.3:1) and was als
statistically significant.We found no statistically significant difference between the ratios of etiology in the two groups.According to the revised Atlanta classification, a slight increase in the moderate pancreatitis cases and a slight decrease in the mild cases in the peri-COVID group without statistical significance were observed.The mean length of hospitalization was statistically significantly shorter in the peri-COVID group; however, the number of complications and the mortality were statistically significantly higher in the peri-COVID group (Table 2).

Comparing the data of the AP patients with the COVID-19 infection (the COVID-19 group) with those with non-COVID-19 AP in the peri-COVID group, the mean age was statistically significantly lower in the COVID-19

ratio of alcoholic and unknown etiology was statistically significantly increased compared
o the non-COVID-19 patients, for whom biliary etiology also fell below half.The average duration of hospitalization was similar in the two groups but with a statistically significant increase n the rate of complications with surgical risk, which almost doubled in the COVID-19 group, with a halving of the operability.Mortality was almost three times higher in the group of patients with AP and COVID-19 compared to the non-COVID-19 group (Table 3).Regarding the variation of the biological parameters in the complications of the patients in the pre-COVID group, the values of the group of patients with complications and the group of patients without complications were compared, and statistically significant variations were recorded only among the lymphocytes (1.16 ± 0.39 vs. 1.72 ± 0.79, p < 0.001).

In the peri-COVID group, following the comparison of the values from the group of patients with complications and the group of patients without complications, statistically significant variations were recorded only among the MCVs (99.63 ± 5.12 vs. 92.02± 7.31, p < 0.001).

Regarding the variation of the biological parameters in the mortality of the patients in the pre-COVID group, the values of the group of patients who died were compared with the group of patients who survived, and statistically significant variations were recorded among the neutrophils (14.38 ± 7.48 vs. 10.41 ± 5.53, p = 0.002), lymphocytes (0.91 ± 0.41 vs. 1.74 ± 0.94, p < 0.001),

0.005) (Figure 2).* p < 0.05-statis
ically significant; NLR-neutrophil-lymphocyte ratio, PLR-platelet-lymphocyte ratio, MLR-monocyte-lymphocyte ratio, dNLR-derived neutrophil-lymphocyte ratio, AISI-aggregate index of systemic inflammation, SIRI-systemic inflammatory response index, SII-systemic immune-inflammation index, MCVL-average corpuscular volume-lymphocyte ratio, IIC-cumulative inflammatory index.


The Predictive Values of Inflammatory Indices in Mortality

The prediction of the risk of death in the pre-COVID group using the ROC curve showed the following significant results: the cutoff value of the NLR level was found to be >11.01 with 91.7% sensitivity and 80.5% specificity (AUC = 0.833, p < 0.001); the cutoff value of the MCVL level was found to be >72.14 with 91.7% sensitivity and 67.6% specificity (AUC = 0.817, p < 0.001); and the cutoff value of the IIC level was found to be >13.29 with 91.7% sensitivity and 78.6% specificity (AUC = 0.887, p < 0.001) (Figure 4).The prediction of the risk of death in the peri-COVID group using the ROC curve showed the following (Figure 5): the cutoff value of the NLR level was found to

tivity and 4
.5% specificity (AUC = 0.743, p < 0.001); PLR, MLR, dNLR, AISI, SIRI, SII had the area < 0.700; the cutoff value of the MCVL level was found to be >74.9 with 94.4% sensitivity and 57.5% specificity (AUC = 0.762, p < 0.001); and the cutoff value of the IIC level was found to be >12.12 with 91.7% sensitivity and 72.5% specificity (AUC = 0.870, p < 0.001) (Table 8).The prediction of complications in the peri-COVID group using the ROC curve showed the following significant values (Figure 3): MCVL was found to be >78 with 80% sensitivity and 60.3% specificity (AUC = 0.681, p < 0.001), and IIC was found to be >10.51 with 72.5% sensitivity and 52 6% specificity (AUC = 0.686, p < 0.001) (Table 7).ficity (AUC = 0.817, p < 0.001); and the cutoff value of the IIC level was found to be >13.29 with 91.7% sensitivity and 78.6% specificity (AUC = 0.887, p < 0.001) (Figure 4).The prediction of the risk of death in the peri-COVID group using the ROC curve showed the following (Figure 5): the cutoff value of the NLR level was found to be >5.93 with 91.7% sensitivity and 42.5% specificity (AUC = 0.743, p < 0.001); PLR, MLR, dNLR, AISI, SIRI, SII had the area < 0.700; the cutoff value of the MCVL level was found to be >74.9 with 94.4% sensitivity and 57.5% specificity (AUC = 0.762, p < 0.001); and the cutoff value of the IIC level was found to be >12.12 with 91.7% sensitivity and 72.5% specificity (AUC = 0.870, p < 0.001) (Table 8).* p < 0.05-statistically significant.NLR-neutrophil-lymphocyte ratio, PLR-platelet-lymphocyte ratio, MLR-monocyte-lymphocyte ratio, dNLR-derived neutrophil-lymphocyte ratio, AISI-aggregate index of systemic inflammation, SIRI-systemic inflammatory response index, SII-systemic immune-inflammation index, MCVL-average corpuscular volume-lymphocyte ratio, IIC-cumulative inflammatory index.


Univariate and Multivariate Analysis of Predictive Factors for Complications

The univariate analysis of the inflammatory markers was performed using the optimal cutoff obtained after the ROC curve analysis.

The univariate analysis results indicated that in the pre-COVID group there was a significant association of the complications with the lymphocytes: PLR > 97.3, MLR > 0.66, SIRI > 3.56, MCVL > 64.89, and IIC > 8.41.In the multivariate analysis, the value of the lymphocytes (p = 0.032) and of the inflammatory markers MCVL > 64.89 (OR 3.52, 95% CI 1.52-8.13,p = 0.003), and IIC > 8.41 (OR 5.56, 95% CI 2.87-10.95,p = 0.049) maintained a significant association with the complications (Table 9).In the peri-COVID group, in the univariate analysis, the complications were associated with the area, proteins, ALT, MCV, MCVL > 78, and IIC > 10.51.The multivariate analysis showed a significant association of the complications with the area (p = 0.027) and ALT (p = 0.026) and among the inflammatory markers only with MCVL > 10.51 (OR 4.22, 95% CI 1.46-12.14,p = 0.008) (Table 10).


Univariate and Multivariate Analysis of Predictive Factors for Mortality

The univariate analysis indicated that in the pre-COVID group mortality was associated with levels of urea, BT, neutrophils, lymphocytes, NLR > 11.01, PLR > 102.88, dNLR > 3, SIRI > 4.26, SII > 906.21,MCVL > 72.14, and IIC > 13.29.The multivariate analysis showed a significant association of mortality with the levels of urea (p = 0.013), BT (p = 0.009), neutrophils (p = 0.025), and lymphocytes (p = 0.013) and among the inflammatory markers with NLR > 11.01 (OR 20.10, 95% CI 3.12-129.42,p = 0.002) and IIC > 13.29 (OR 18.71, 95% CI 2.60-134.52,p = 0.004) (Table 11).In the peri-COVID group, in the univariate analys s, mortality was associated with protein levels, urea, creatinine, BT, INR, COVID-19, Hb, Ht, neutrophils, lymphocytes, RDW, NLR > 5.39, MCVL > 79.4, and IIC > 12.12.The multivariate analysis showed a significant association of mortality with the level of protein (p = 0.009), urea (p = 0.015), creatinine (p < 0.001), BT (0.003), infection with COVID-19 (0.001) and neutrophils (0.002) and among the inflammatory markers with NLR > 5.3 (OR 10.24, 95% CI 1.29-81.17,p = 0.028), MCVL > 79.4 (OR 8.92, 95% CI 5.21-141.58,p = 0.041), and IIC > 12.12 (OR 27.94, 95% CI 3.57-218.58,p = 0.002) (Table 12).


Results of Pearson Chi-Square Test

According to the results obtained in the multivariate analysis, we performed the Pears n chi-square test to evaluate the correlation between the MCVL and the complications with surgical risk and the IIC with mortality, which had a value of p < 0.001 in both cases.We also calculated for the MCVL the risk of developing complications, which is 1.85 times in the pre-COVID group and 0.37 times in the peri-COVID group at the cutoff values of 64.89 and 78, respectively.For IIC, the risk of death was 4.33 times in the pre-Covid group and 3.33 times in the peri-Covid group at the cutoff values of 13.29 and 12.12, respectively (Table 13).

Table 13.Chi-square analysis of the risk of complications and mortality of IIC and MCVL in the pre-COVID and peri-COVID period and the adjusted cutoff value of the IIC in the pre-COVID group.


Pre-COVID


Discussions

The present study investigates the predictability of surgical risk complications and mortality in patients with AP and the influence of the SARS-CoV-2 infection on it.For this, we used i flammation markers that were studied separately in the literature, without comparative studies before and during the COVID-19 pandemic.Moreover, in addition to the existing inflammatory markers, we introduced two new inflammation markers that better reflect the inflammation-induced changes to white and red cells in the peripheral blood.Wang et al. were the first to describe nine cases of acute pancreatitis in a series of 52 patients with COVID-19 [26].

The pandemic led to global chaos, particularly in the healthcare system where it caused sudden interruptions in the provision of healthcare to all patients across the country, caused by the relocation of workforces from healthcare facilities to treating patien s with COVID-19 [27] or the panic instilled in the population that led to a decrease in trust in the medical act with a therapeutic [28] or in a preventive purpose such as vaccination [29].The number of non-COVID-19 acute emergency cases decreased, a fact reflected by the low number of emergency presentations and the number of abdominal CT scans performed [30].Moreover, a negative impact of COVID-19 was found in patients who had risk factors for diseases or already suffered from life-threatening diseases such as AP, with some hesitation from the patient to initially see a doctor because of the associated risk of exposure to the virus [31], a condition also found in our study, with 196 patients with AP vs. 237 who presented themselves in the pre-COVID period.There was also a decrease in the number of patients who presented with a mild form, from 34% to 19.8% according to the Atlanta classification, fact explained by the fact that sometimes the patients resorted to self-medication by using the personal medicines of another family member prescribed by a doctor for chronic or recurrent diseases or symptoms [27].There was a slight increase in the mild and severe forms and an increase in the time elapsed between the onset of symptoms and hospital presentation, resulting in an increase in the disease severity, ultimately causing th patient to present himself in the hospitalization unit with a more severe form of the disease.

The infection with COVID-19 affects the lungs as the main organ of the infection, but in cases with mild, moderate, and severe infection, gastrointestinal signs and symptoms were also reported, which included nausea, diarrhea, anorexia, abdomina pain, belching and vomiting [32], which led to some cases already hospitalized with COVID-19 with digestive symptoms refractory to treatment to be diagnosed with AP, which was classified as being of unknown etiology.In the peri-COVID group, a slight increase in the number of cases of PA of unknown tiology was observed, and more than half of them were patients in the COVID-19 group.

Acute pancreatitis is a common disease that has a sudden onset, rapid progression, and high mortality and morbidity; therefore, as soon as patients are diagnosed with this disease, their condition should be monitored as quickly as possible and with maximum accuracy.The need to anticipate their prognosis must lead to a treatment of the real suffering, which is sometimes masked by the associated conditions.As inflammatory mediators play a significant role in the occurrence of PA, various inflammatory markers were recently used to assess the predictive value of the severity and mortality of patients with PA.

In our study, we investigated the value of the known markers NLR, PLR, MLR, dNLR, AISI, SIRI, and SII and the newly introduced MCVL and IIC as independent predictive factors of surgical risk complications and mortality in patients with AP; we tested their accuracy in two time periods, namely before COVID-19 and during COVID-19.Their cutoff points for complications with surgical risk and mortality were observed using the ROC curve.

We did not use other serum predictive markers such as C-reactive protein, procalcitonin, interleukin-6, and interleukin-8 because there are numerous studies in which they cannot accurately predict the prognosis and se erity of PA [13], and they were not used in our case on a wide scale.

Complications with surgical risk that can put life at risk are abscess, necrosis, and pancreatic pseudocyst.The pancreatic abscess is a complication that occurs particularly in patients with acute pancreatitis and is defined as a circumscribed intraabdominal collection of pus that is usually located near the pancreas and is caused by bacterial colonic translocation [33].In AP patients whose abdominal symptoms do not improve after drug treatment and who begin to develop systemic inflammatory response syndrome, intraabdominal infection should be suspected and may require surgical treatment depending on the patient's condition.In our case, a doubling of the incidence of abscesses, from 3.8% to 7.7%, was observed in the peri-COVID period with a higher incidence among COVID-19 patients, which was probably explained by their increased susceptibility to secondary infections, a fact reported in several reviews and meta-analyses [34,35] (Table 3).

Pancreatic necrosis is one of the serious complications of AP, with a significant proportion of these patients requiring surgical intervention.The main cause of death in a patient with pancreatic necrosis is infection of the necrotic tissue, which is a poor prognostic factor [36]; thus, pancreatic necrosis without infection has a mortality rate of approximately 15%, while patients with infected necrosis have a mortality rate of between 30 and 39% [37].In our study, an increase in the number of cases of pancreatic necrosis was observed from 3.8% to 10.2%, a fact explained by the doubling of the incidence among patients with COVID-19 and PA, from 8.9% to 17.9%, possibly caused by the severity of these cases.

A pancreatic pseudocyst is a localized fluid collection that is rich in amylase and oth r pancreatic enzymes and is surrounded by a wall of fibrous tissue that is not lined by epithelium [38].The treatment of a pancreatic pseudocyst can consist of open surgical drainage, endoscopic drainage, and percutaneous drainage.Surgical treatment is still used as a treatment option and involves internal drainage in the stomach, duodenum, and jejunum [39].Endoscopic drainage did not show any significant difference between the success rates of surgical and endoscopic treatment, adverse events, and pseudocyst recurrence [40].Percutaneous therapy showed a higher failure rate, a higher mortality and morbidity rate, and patients required more days of long-term hospitalization compared to those treated with surgery [41].In our study, the incidence of the pancreatic pseudocyst was higher in the pre-COVID group.

Although, overall, the rate of complications with surgical risk was higher in the peri-COVID group, 68.7% occurred in patients in the COVID-19 group, with a rate of surgery reduced by half due to the contraindication of performing s rgery in the context of already established multiorgan failure.Cases of severe AP and COVID-19 are characterized by a cytokine storm that ultimately leads to multiorgan failure and increased mortality.

Liu et al. studied the damage to the pancreas by the SARS-CoV-2 infection and the association with ACE-2 receptors, which are slightly more expressed in the pancreas than in the lungs, possibly leading to the direct damage to the pancreas that was reported in 2% of non-severe patients, compared to 17% of patients with severe COVID-19 [12].

The global mortality of AP is approximately 10-15%, reaching 30-50% in severe AP [1,42].In our study, we recorded a mortality of 10.1% in the pre-COVID group, with an increase to 18.4% in the peri-COVID group, where a mortality of 42.9% was recorded among patients with a COVID-19 infection and AP.

The neutrophil-lymp ocyte ratio (NLR) reflects the systemic inflammatory state and has a prognostic value in cardiovascular diseases, autoimmune diseases, tumors, and other infectious diseases [43][44][45].Although, until now, the exact limit for NLR was missing, Forget et al. established that the normal values of NLR in a healthy, non-geriatric adult population is between 0.78 and 3.53 [46].The first study that analyzed the relationship between NLR and the outcomes in patients with acute pancreatitis was conducted by Yao J. [47] in 2011.In a study by Jeon and Park, it was shown that high basic NLR was associated with severe PA and organ failure [48].Azab et al. found that NLR was increased in PA patients, and they also reported that NLR was superior to total WBC or individual neutrophil or lymphocyte counts in predicting ICU admission and mortality for PA [49].In another study that assessed the severity of PA, the NLR cutoff was >4.7, with a sensitivity o 90% and a specificity of 22% [13] In this study, regarding the constitution of the complications with surgical risk in the pre-COVID vs. peri-COVID period, the NLR had a cutoff of >5.58 vs. 6.41, with a sensitivity and specificity of 78.6% and 42.2% vs. 80% and 44.9%.The prediction of the complications was significant, but the specificity in both groups remained below 50%.

Regarding mortality, in the pre-COVID vs. peri-COVID period the NLR had a cutoff >11.01 vs. 5.93, with a sensitivity and specificity of 91.7% and 80.5% vs. 91.7%and 42.5%.The prediction of mortality was statistically significant in both groups, bu the specificity dropped dramatically in the peri-COVID group, below 50%, with a halving of the cutoff value, making it an inapplicable marker in groups with impaired immunity due to other superimposed infectious conditions with PA.

In recent years, studies have shown that platelets and lymphocytes play a critical role in the inflammatory process; therefore, in recent research, the platelet lymphocyte ratio (PLR), which is a new inflammatory factor, received research attention, proving that it can be a good indicator of inflammation in multiple acute cardiac and renal diseases [50] and in acute pancreatitis [36].Thrombocytopenia is related to ch onic liver disease due to the impaired production of platelets and the decreased hepatic synthesis of thrombopoietin [51]; therefore, PLR can vary depending on liver function, but also on systemic inflammation.Increased NLR and PLR values have been associated with inflammatory conditions, and poor results in severe PA are explained by uncontrolled SIRS and its progression to multiple organ dysfunction syndrome [13].Regarding the constitution of the complications with surgical risk in the pre-COVID vs. peri-COVID period, PLR had a cutoff of >97.3 vs. 140.2,with a sensitivity and specificity of 89.3% and 34.5% vs. 72.5% and 50.6%.The prediction of the complications was significant, but the specificity was low in the pre-COVID group, and the differences between the cutoff values were large.In terms of mortality, in the pre-COVID vs peri-COVID period PLR had a similar cutoff but with a specificity below 40%.

The monocyte to lymphocyte ratio (MLR) is the absolute number of monocytes divided by the absolute number of lymphocytes and is a hematological and inflammatory parameter.The MLR is associated with var ous infectious and inflammatory conditions as well as with the systemic inflammatory response, which reflects the immune status of the diseases [52,53].The monocyte-lymphocyte ratio (MLR) was associated with adverse outcomes in colorectal and urologi al cancers [54,55].Regarding the formation of complications with surgical risk in the pre-COVID vs. peri-COVID period, the MLR had a cutoff of >0.66 vs. 0.53, with a sensitivity and specificity of 78.6% and 69.9% vs. 75% and 49.4%.The prediction of complications was significant, but the specificity was slightly decreased in the peri-COVID group, without large differences between the cutoff values, which makes it a faithful marker in the prediction of complications.Regarding mortality, in the pre-COVID vs. peri-COVID period the MLR had a similar cutoff, but with a specificity below 40%.

The de ived neutrophil lymphocyte ratio (dNLR) was studied in the evaluation of the prognosis of patients with metastatic disease regardless of the treatment followed [56], proving that it has a prognostic value similar to that of the classic NLR.It reflects the increase in the number of neutrophils and a relative decrease in the number of lymphocytes, suggesting that inflammatory reactions are dependent on neutrophils.Regarding the formation of complications with surgical risk in the pre-COVID vs peri-COVID period, the dNLR predicted the formation of complications and mortality, but with a specificity below 50%.AISI, SIRI, and SII are considered to reflect the immune and inflammat ry status, and a link between them and the risk of mortality in different types of cancer [57], autoimmune diseases [58], and infectious diseases [21] has been observed.

The aggregate index of systemic inflammation (AISI) had statistical significance in the prediction of complications only in the pre-COVID group, but with a very low specificity of less than 10%.The AISI also predicted mortality only in the pre-COVID group, but with a specificity below 50%.

The systemic inflammatory response index (SIRI) had statistical significance only n both groups with regard to the prediction of complications and mortality, but with a sensitivity below 50%.

The systemic immuno-inflammatory index (SII), derived from neutrophils, lymphocytes, and platelets, was used for the first time in 2014 in evaluating the prognosis of hepatocellular carcinoma [59].A recent study showed that the SII was more sensitive than the NLR and PLR in distinguishing between mild and severe PA [20].The SII was statistically significant in predicting complications and death only in the pre-COVID group, but with a sensitivity below 50%.

Red c lls (RBC), which are produced in the bone marrow, are the most common type of cells in the blood [60] and have the function of supplying oxygen to organs and tissues.The MCV is the average volume of an RBC and is not only an indicator of anemia but is also a marker of chronic inflammation [61], being found to be assoc ated with heart failure, diabetes, vascular accident, and venous thromboembolism [62].Moreover, an increased MCV is an indicator of the deterioration of liver function, with macrocytosis being found in various liver diseases that lead to changes in the lipid composition of the RBC membrane [63].

The impact of a massive increase in the relative mass of RBC is an increase in the viscosity of whole blood, primarily by increasing the number of particles per unit of blood volume and, therefore, increasing the peripheral resistance to blood flow [64].Viscosity is defined as the ratio between the shearing effort and the shear rate and is a risk factor for deep venous thrombosis, creating a procoagulant state through thrombocytosis and increasing the concentration of the circulating tissue factor 65]. Mean corpuscular volume is considered to have qualitative effects on blood rheology, rather than quantitative effects, such as hematocrit [66].

MCV alone has no role as a biomarker of severity in patients with AP, although it was found to have a value greater than 100 in the

atients who died in the p
ri-COVID group.Its correlation with lymphocytes, whose low value shows a diminished initial immune response of the host [67], led to the establishment of a new prognostic marker of the severity of patients with PA, which we called the MCVL ratio, which was obtained by dividing the absolute number of the erythrocyte MCV by the absolute number of lymphocytes.Regarding the occurrence of complications with surgical risk in the pre-COVID vs. the peri-COVID period, the MCVL had a cutoff of >64.89 vs. >78, with a sensitivity and specificity of 78.6% and 56.8% vs. 80% and 60.3%.The prediction of complications was signifi

nt, with the
best values among all the studied markers.The MCVL also proved its superiority in predicting mortality in the pre-COVID vs. peri-COVID period with a cutoff of >72.14 vs. 74.9, with a sensitivity and specificity of 91.7% and 67.6% vs. 91.7%and 57.3%; these are close values, which makes it useful in both types of groups.

Regarding the prediction of complications in the peri-COVID group, the multivariate analysis showed that the MCVL threshold value of 78 is an independent factor for the prediction of complications, with a slight decrease in the OR from 4.84 to 4.22.

The erythrocyte distribution width parameter (RDW) (the coefficient of variation of the RBC volume) is calculated after a routine complete blood count (CBC) and is the ratio between the standard deviation of the RBC volume and the mean corpuscular volume, which is useful in quantifying the heterogeneity of the RBC volume (anisocytosis).In addition to the association with a series of cardiovascular diseases, cancer, and diabetes [68], studies have reported an increased RDW in critically ill patients, making