Prevalence and Outcomes of Pancreatic Enzymes Elevation in Patients With COVID-19: A Meta-Analysis and Systematic Review

Background: Although coronavirus disease 2019 (COVID-19) is considered to be a disease that mainly involves the respiratory system, an increasing number of studies have reported that COVID-19 patients had pancreatic enzymes (PE) elevation and even pancreatic injury. The study aims to determine the prevalence of PE elevation, and the relationship between elevated PE and prognosis in COVID-19 patients. Methods A comprehensive literature search was conducted according to the PRISMA guideline in PubMed, Embase, Scopus, Web of Science, and Google Scholar for studies reporting PE elevation in patients with COVID-19 from 1st January 2020 to 24th November 2021. Results A total of 13 studies (24,353 participants) were included in our review. The pooled prevalence of PE elevation in COVID-19 patients was 24% (18%–31%), the pooled odds ratio (OR) of mortality was 2.5 (1.7–3.6), the pooled OR of ICU admission was 4.4 (2.8–6.8), and the pooled OR of kidney injury, respiratory failure and liver injury were 3.5 (1.6–7.4), 2.0 (0.5–8.7), and 2.3 (1.4–3.9) respectively. In addition, the subgroup analysis revealed that although PE elevated to > 3 × upper normal limit (ULN) was significantly related to the mortality (OR = 4.4, 2.1–9.4), it seemed that mild elevation of PE to 1–3 ULN also had a considerable risk of mortality (OR = 2.3, 1.5–3.5). Conclusions PE elevation was a common phenomenon in patients with COVID-19, and was associated with poor clinical outcomes. However, due to the limited numbers of included studies, the result of our study still needed to be validated. Systematic Review Registration https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=295630, identifier: CRD42021295630.


INTRODUCTION
Coronavirus disease 2019 (COVID-19) is a novel severe respiratory infectious disease caused by severe respiratory syndrome coronavirus-2 (SARS-CoV-2). Since the first case was officially reported in Wuhan, China in December 2019, COVID-19 has experienced a widespread outbreak and epidemic worldwide, which has caused tremendous impact and pressure on the medical and health systems around the world (1). On March 11, 2020, the World Health Organization announced it as a global pandemic disease. As of November 28, 2021, over 260 million confirmed cases have been reported globally, of which nearly 5.2 million died (2). COVID-19 has now developed into a global health crisis.
Although SARS-CoV-2 was believed to mainly invade the respiratory system of patients, with clinical manifestations as fever, cough, shortness of breath, and extensive lung consolidation, it cannot be ignored that some patients simultaneously had digestive symptoms as nausea, vomiting, and diarrhea (3)(4)(5)(6). Consistent with SARS-CoV, SARS-CoV-2 invades cells through combining its spike protein with the angiotensin-converting enzyme II (ACE II) receptors (7,8). Existing studies suggested that, in addition to type II alveolar epithelial cells, ACE II receptors are also highly expressed in esophagus, small intestine, colon and pancreas (9)(10)(11), and show a high affinity for SARS-CoV-2. Therefore, the pancreas may also be a potential target of SARS-CoV-2, which can lead to undetectable pancreatic injury (11).
Wang et al. (12) first reported pancreatic enzymes (PE) elevation in COVID-19 patients in a study involving infected people in the early stage of the epidemic. Subsequently, an increasing number of studies reported the similar findings. Since the critically ill COVID-19 patients often experience severe systemic inflammatory, shock, microcirculatory disturbance and renal failure, some scholars believed that PE elevation might be associated with pancreatic ischemic injury (13)(14)(15), and the elevated PE can serve as a surrogate marker for poor prognosis of COVID-19 patients. However, in different studies, due to the different sample sizes and definition of PE elevation, the prevalence of PE elevation varied greatly, and the clinical significance of it was still controversial (14,16). In a previous meta-analysis by Goyal et al. (17), hyperlipasemia was found to be associated with the severity of COVID-19. However, in their study, severe COVID-19 was defined as death, intensive care unit (ICU) admission and need for mechanical ventilation, which was not rigorous because the elevated PE may have different impacts on different clinical outcomes. In addition, the included studies in their meta-analysis included letter to editor and correspondence, lacking enough casecontrol and cohort studies covering large samples and multicenters. We believed that the result of their study was open to question. Therefore, we performed this meta-analysis and systematic review in order to 1) determine the prevalence of PE elevation in COVID-19 patients, and 2) summarize the impact of PE elevation on the clinical outcomes in patients with COVID-19.

Protocol Registration
This meta-analysis and systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (18), and this study was part of the registered protocol on the International Prospective Register of Systematic Reviews (CRD42021295630).

Eligibility Criteria
Based on the PICOS (Population, Intervention/Exposure, Comparison, Outcome, and Study design) strategy, the inclusion criteria were as follows: Population: participants included in studies were clearly diagnosed with COVID-19.
• Comparison: normal level of PE.
The exclusion criteria were as follows: • Non-adult studies.
• Studies with unavailable full-text.
• Studies with unclear criteria for PE elevation.
• Studies not providing specific prevalence or outcomes of PE elevation. • Protocols, review articles, abstracts, letters to editor, correspondence, case reports, and pre-prints.

Study Selection
All identified articles were first imported into the Endnote X9 software to remove duplicates manually, then the titles and abstracts of studies were screened by two reviewers (XXY and QC) blindly in accordance with the inclusion and exclusion criteria to exclude irrelevant articles. The articles meeting the eligibility criteria were next screened on full text by the same two reviewers. Any disagreements were resolved by consulting another reviewer (YZ).

Data Extraction
Data were extracted by two reviewers (YZ and XXY) using a designed Excel sheet. Any disagreements were solved by another reviewer (LHH). The following information was recorded: 1) author, 2) year of publication, 3) country, 4) study type, 5) samples size, 6) type of elevated PE, 7) definition for PE elevation, 8) proportion of patients with PE elevation among all patients, 9) proportion of patients with acute pancreatitis (AP) among patients with PE elevation, 10) clinical outcomes of COVID-19 patients with PE elevation.

Quality Assessment
The Quality in Prognostic Studies tool was used to assess the quality of the included studies, which includes six items: study participation, study attrition, prognostic factor measurement, outcome measurement, confounding measurement and account, and analysis (19). Each article was assessed by two reviewers (YZ and QC) independently using a consistent standard. Any disagreements were resolved by consulting another reviewer (GTL).

Statistical Analysis
The statistical analysis was performed using the Stata SE Version 16 software. We conservatively used a random-effects model to analyze the impact of PE elevation on mortality, ICU admission, and complications. A forest plot was used to visualized the data. The heterogeneity of included studies was estimated using the Cochran's Q-test and I 2 statistics, and the value of I 2 between 0 and 25%, 25-75%, and >75% was considered mild, moderate, and high heterogeneity, respectively (20). Prespecified subgroup analyses based on the definition of PE elevation and data source were performed to explore the heterogeneity of clinical outcomes between studies. Sensitivity analyses were preformed to explore the impact of each study by removing studies one by one. Egger's test and visual inspection of funnel plot were used to examine the publication bias. A P-value <0.05 was considered statistically significant.

Search Results
The PRISAM flow diagram showed the process of article selection (Figure 1). A total of 1,538 records were extracted from the initial search, and three additional studies were identified through the reference searching. After removing the duplicates (n = 421), we screened 1,120 studies with titles and abstracts, of which 76 studies meeting the eligibility criteria were reviewed with full text. Thirteen studies (12-14, 16, 21-29) were finally included for qualitative and quantitative analysis in this review. Table 1 summarized the characteristic of the included studies. Thirteen studies were from the USA (n = 5), China (n = 3), Italy (n = 2), Turkey (n = 2), and Germany (n = 1), of which, 12 were retrospective, one (24) was prospective, and five studies (14,16,23,25,29) were multicenter. The sample size ranged from  (Figure 2). Six studies (13,14,21,22,25,26) with samples more than ten patients reported AP diagnosis according to the revised Atlanta classification of acute pancreatitis 2012 (30)   9% (95% CI: 2%-15%), with a high degree of heterogeneity (I 2 = 93.2%) (Supplementary Figure 1).

Analysis of PE Elevation and Mortality
Ten studies (13, 14, 16, 21-23, 25-27, 29) reported the mortality associated with elevated PE. A total of 2,4207 participants including 3,142 participants in the elevated PE group, of which 760 participants died, and 2,1065 participants in the normal PE group, of which 2,033 participants died were included in the analysis. The mortality ranged from 10.0 to 79.3%. PE elevation was significantly related to the mortality of COVID-19 patients (OR = 2.5, 95% CI: 1.7-3.6), with substantial heterogeneity (I 2 = 89.5%) (Figure 3).
Since the heterogeneity was significant, we performed a sensitivity analysis to explore the impact of each study. The result showed that two studies (13,23) affected the pooled OR (odds ratio) of mortality (Supplementary Figure 2). After removing any one of the two studies, the I 2 did not decrease significantly (78.6-87.7%). After removing both studies simultaneously, the heterogeneity became acceptable (I 2 = 13.3%), and the pooled OR was 1.7 (95% CI: 1.5-1.9) (Supplementary Figure 3).

Analysis of PE Elevation and Hospitalization
Six studies (14,16,21,(26)(27)(28) reported PE elevation was associated with ICU admission in COVID-19 patients. A total of 1,783 participants including 520 participants in the elevated PE group, of which 147 were admitted to the ICU, and 1,263 participants in the normal PE group, of which 138 were admitted to the ICU were included in the analysis. As is shown in Figure 4, PE elevation was significantly associated with ICU admission in COVID-19 patients (OR = 4.4, 95% CI: 2.8-6.8), with acceptable heterogeneity (I 2 = 36.8%).
Three studies reported that PE elevation was related to the length of stay (LOS). Ahmed et al. (14) reported that the median LOS for patients with elevated PE was 15 days (IQR: 8.3-30 days), and that for patients with normal PE was 11 days (IQR: 5.5-20.5 days). Akkus et al. (27) found that the median LOS for patients with elevated PE was 11.5 days (range: 3-41 days), and that for patients with normal PE was 8 days (range: 0-38 days). Benias et al. (29)

Publication Bias
Egger's test revealed that, there was no significant publication bias for studies reporting mortality (P = 0.463), ICU admission (P = 0.647), and KI (P = 0.523) associated with PE elevation, except for PE prevalence (P = 0.006). Supplementary Figure 6 for visual funnel plots.

DISCUSSION
To the best of our knowledge, this is the latest and most comprehensive systematic review and meta-analysis on the prevalence and clinical outcomes of PE elevation in COVID-19 patients. Our study demonstrated that, overall, PE elevation was common in COVID-19 patients. The pooled prevalence of PE elevation was 24%, which was significantly higher than that of the previous mate-analysis by Goyal et al., and the risk of developing severe COVID-19 in patients with hyperlipasemia in their study was higher than that of adverse outcomes in our study (17). This is understandable because their metaanalysis included fewer and earlier studies. In addition, we also found that about 9% of patients with elevated PE > 3× ULN eventually developed AP, which was also higher than that of a previous meta-analysis on the prevalence and clinical outcomes of AP in COVID-19 patients reported by Yang et al. (31). In their study, the pooled prevalence of AP complicated by COVID-19 was about 3.1%, of which about 18.5% eventually died. COVID-19 patients with pancreatic injury often had poor clinical outcomes. According to the revised Atlanta classification of acute pancreatitis 2012, the diagnosis of AP included abdominal pain, the elevated PE > 3× ULN, and characteristic findings of AP on imaging. However, it cannot be denied that patients did not meet the diagnostic have no potential pancreatic injury and potential risk of poor prognosis. In this systematic review and meta-analysis, we extensively searched and included existing studies on PE elevation in patients with COVID-19 and included more participants to reveal the association between elevated PE and the clinical outcomes of COVID-19 patients. At present, the cause of PE elevation was still unclear. In the autopsy of patients with severe acute respiratory syndrome, SARS-CoV was found to be present in pancreatic tissue (32). Due to the similarity of the two viruses and the ACE II receptors highly expressed in the pancreas, pancreatic injury caused by the direct invasion of SARS-CoV-2 was one of the potential causes of PE elevation. In the case report by Schepis et al., SARS-CoV-2 RNA was detected for the first time in a pancreatic pseudocyst fluid sample from a COVID-19 patient (33). In addition, a Chinese pathology study found that COVID-19 patients had a small amount of pancreatic islet cell degeneration (34). Although the above studies seemed to verify the possibility of direct damage by SARS-CoV-2, in critically ill patients, PE elevation often occurred. The most widely accepted explanation for PE elevation with non-viral causes was pancreatic ischemia (15,35,36). When the patient had severe infection, hypoperfusion and shock, the pancreas was insufficiently perfused, which will lead to pancreatic injury. In addition, non-pancreatic causes such as intestinal inflammation (37), diabetes (38), acidosis (39), and renal failure (38,40,41) can also lead to PE elevation. Although a variety of causes, including pancreatic injury, can lead to PE elevation, it was undeniable that when the above symptoms appeared in COVID-19 patients, it often indicated the occurrence of poor clinical outcomes.
Our meta-analysis and systematic review found that PE elevation in COVID-19 patients was significantly associated with the increase of mortality, ICU admission, LOS, and clinical complications as KI, respiratory failure and liver failure. In the analysis of mortality, the pooled OR without two heterogeneous studies (13, 23) was 1.7, which was significantly lower than the pooled OR of 2.5 for all studies. In these two studies, we found that 50.1% of the patients were diagnosed with severe COVID-19 and 43% with severe pneumonia, respectively. In a meta-analysis involving 30 studies, the proportion of severe COVID-19 was about 26% (42). In addition, studies have shown that severe COVID-19 and more comorbidities were the risk factors for higher clinical mortality (43,44). Although the severity of COVID-19 was not clearly reported in other included studies, we believed that the heavier condition of patients contributed to the higher mortality, resulting in the overestimation of the risk of pooled mortality. Compared with the existing discovered risk factors for mortality of COVID-19 patients such as gender (45,46), age (43,(45)(46)(47), diabetes (45,48), history of COPD (45), and chronic cardiac disease (49), we found that PE elevation had a similar risk for mortality. Therefore, PE elevation may also serve as a risk indicator of mortality for patients with COVID-19.
In the subgroup analysis of mortality based on different definition of PE elevation, we found that patients with elevated PE of >3× ULN had a higher risk of death. In addition, it was also interesting that regardless of including or excluding the heterogeneous studies, the pooled OR of mortality in the 1-3 ULN group did not change a lot (2.3 vs. 2.5, 1.7 vs. 1.7), which indicated that a slight increase in PE, even if it did not satisfy the diagnostic criterion of AP, will have a hazardous effect on the clinical outcomes of COVID-19 patients. In other words, it is possible that PE is a sensitive marker for predicting the mortality in COVID-19 patients.
In the subgroup analysis on mortality based on different data sources, the pooled OR of mortality in single-center studies was higher than that of multi-center studies (2.4 vs. 1.6). Among the included multi-center studies, one was a database study (25), one study was based on two tertiary hospitals and four community hospitals (16), and two studies was based on several major tertiary medical systems (14,29). Since the time of data recorded and the methods of measurement and testing in public databases were difficult to ensure consistency (50), and the conditions of patients in community hospitals were different from those in tertiary medical institutions, we believed that the existing multicenter studies may underestimate the real risk of mortality in COVID-19 patients with elevated PE. Therefore, we hoped that prospective studies based on several tertiary medical institutions can be carried out to explore the real risk of hospital mortality related to PE elevation in COVID-19 patients. And on this basis, further explore the specific sources and risk factors of PE elevation.
In addition, it is worth noting that Ahmed et al. (14) tried to explore the relationship between PE elevation, D-dimer and mortality, ICU admission. Existing studies have proved that laboratory factors including D-dimer levels, demographic factors, patient history factors, physical examination factors, and clinical scores were significantly related to the severity and poor prognosis of COVID-19 patients (51). Since the COVID-19 patients often underwent various examinations during hospitalization, which generated rich medical records, it will be a meaningful attempt to predict the clinical outcomes of COVID-19 patients through using these multi-dimensional data. At present, machine learning algorithm has been widely used in the prediction tasks of complications, mortality, etc. in COVID-19 (52)(53)(54). We hoped that future studies can develop similar predictive models based on multi-omics clinical data including PE elevation to predict the clinical outcome of COVID-19 patients.
This systematic review and meta-analysis also had certain limitations. First, we only searched articles in English, which may lead to potential bias of publication. Second, due to the limited number of articles included, the results showed significant heterogeneity. Although we attributed it to the differences in the severity of COVID-19, there may also be other potential factors that we overlooked. Third, although we tried to perform a subgroup analysis to explore the impact of PE elevation on specific clinical outcomes, due to the few studies reporting detailed complications, the result of our analysis was unstable and needed to be validated by including more studies in the future.

CONCLUSION
In conclusion, our research found that PE elevation was a risk factor for poor clinical outcomes in patients with COVID-19. Compared with patients with normal PE, patients with elevated PE had a higher risk of mortality, ICU admission, and complications. In addition, future studies are still needed for further analysis of more impacts of PE elevation in COVID-19 patients.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.