Accepted Article
Article Type: Original Article Risk factors for pancreatic infection in patients with severe acute pancreatitis: An analysis of 163 cases1 Yan Bo ZENG,* Xian Bao ZHAN,* Xiao Rong GUO, Hua Gao ZHANG, Yan CHEN, Quan Cai CAI, Zhao Shen LI Department of Gastroenterology, Changhai Hospital, Second Military Medical University, Shanghai, China *These two authors contributed equally to this work. Correspondence to: Zhao Shen LI, Department of Gastroenterology, Changhai Hospital, Second Military Medical University, 168 Changhai Road, Shanghai 200433, China. Email: [email protected] Conflicts of interest: None. Running title: Risk factors for pancreatic infection
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1751-2980.12150
This article is protected by copyright. All rights reserved.
Accepted Article
ABSTRACT OBJECTIVE: We aimed to identify the risk factors for predicting pancreatic infection in patients with severe acute pancreatitis (SAP). METHODS: One hundred and sixty-three patients with SAP were included and divided into two groups based on the presence or absence of pancreatic infection. Demographic and clinical characteristics, laboratory examination results, complications and treatment modalities of these patients were collected from their medical records. Variables were initially screened by univariate analysis and those with statistical significance were then filtered by multivariate analysis to determine the independent risk factors for pancreatic infection in SAP. RESULTS: Patients having SAP with pancreatic infection had lower partial pressure of arterial CO2, peripheral white blood cell count and alkaline phosphatase levels, as well as a higher computed tomography severity index (CTSI) than those without pancreatic infection, while their lactate dehydrogenase levels and partial pressure of arterial O2 (PaO2) were much higher. Pancreatic infection also occurred more commonly in patients receiving late fluid resuscitation than in those who received early fluid resuscitation.Multivariate analyses revealed that increased lactate dehydrogenase level, high CTSI, delayed fluid resuscitation and low PaO2 were independent risk factors for pancreatic infection in SAP. The sensitivity, specificity, positive and negative predictive values for a model combining the parameters in predicting pancreatic infection were 84%, 97%, 88%, and 96%, respectively, with a cut-off value of 0.393, and the AUROC was 0.923. CONCLUSION: Increased lactate dehydrogenase, high CTSI, delayed fluid resuscitation and hypoxemia are independent risk factors for predicting pancreatic infection in patients with SAP. Keywords: acute necrotizing pancreatitis, diagnosis, hypoxemia, pancreatitis, pancreatic infection, risk factors
This article is protected by copyright. All rights reserved.
Accepted Article
INTRODUCTION A sudden inflammation of the pancreas, known as acute pancreatitis (AP), may variably affect adjacent tissues and even remote organs.1 Severe acute pancreatitis (SAP) has been reported to occur in less than 30% of the AP cases, however, it is associated with a high mortality rate.2,3 SAP usually manifests as local complications such as pancreatic necrosis, pseudocyst or abscess, and/or advanced organ failure including renal, cardiocirculatory and pulmonary dysfuction as well as sepsis, etc.4,5 Multiple organ failure (MOF) and severe infectious complications are life-threatening and often associated with pancreatic necrosis. Infected pancreatic necrosis has been found in 30–70% of the patients with necrotizing pancreatitis, and it significantly increases the incidence of systematic complications and is a leading cause for a high mortality in SAP.5–7 Prospective studies8–12 have identified a number of clinical prognostic scores and serum markers that can be used for assessing and monitoring AP. However, the potential of these parameters to identify patients at risk of developing infected pancreatic necrosis remains unknown. Over the past decades therapeutic strategies for SAP have been developed, and conservative therapies now play a prominent role in treating SAP. However, it is still controversial over the use of some modalities such as antibiotic prophylaxis, probiotics and enteral nutrition (EN) for the prevention of pancreatic infection and a decrease of mortality in AP patients.13–17 Diagnosis of pancreatic infection is crucial in determining the therapeutic approaches for SAP. Early diagnosis of and proper therapies for infected pancreatic necrosis are important for improving the prognosis and decreasing the morbidity in such patients. Serum markers that facilitate early diagnosis of pancreatic infection would be clinically useful. In this retrospective study, we aimed to assess the predictive value of a variety of parameters including biochemical markers, scoring systems, complications and treatments for pancreatic infection in patients with SAP treated with conservative therapies, and to identify the independent risk factors for pancreatic infection.
PATIENTS AND METHODS Patients From October 2008 to December 2011, 252 consecutive patients were admitted to Changhai Hospital, Second Military Medical University (Shanghai, China) due to SAP. Of these patients, 163 (96 men and 67 women) aged ≥18 years who had not been treated with probiotics or prophylactic antibiotics before their admission were finally recruited in the study. All these 163 patients were further divided into two groups, that is, pancreatic infection group (n = 32) and non-pancreatic infection group (n = 131). When pancreatic infection was suspected by persisting or new-onset clinical manifestations and/or laboratory results of sepsis after ruled out infection from other sources, pancreatic infection was confirmed by positive results from bacteria culture obtained from computed tomography (CT)-guided or endoscopic ultrasound-guided fine needle aspiration culture. The study was approved by the Institutional Ethics Committee of Changhai Hospital, Second Military Medical University. Data collection Demographic and clinical characteristics of the patients, their laboratory results, complications and treatment modalities were collected by reviewing the medical records of the patients. Clinical data consisted of etiology of SAP, length of hospital stay, mortality, the Acute Physiology and Chronic Health Evaluation (APACHE) II score,18 Ranson’s score9 and CT severity index (CTSI)19. The APACHE II score and the Ranson’s score were determined within This article is protected by copyright. All rights reserved.
Accepted Article
the first 24 h and 48 h after admission, respectively. Contrast-enhanced CT was performed in all the patients and CTSI was determined within 48 h after their admission to assess the extent of pancreatic inflammation and necrosis. Laboratory data were obtained within the initial 24 h from the onset of disease, including pH value, partial pressure of arterial oxygen (PaO2), arterial partial pressure of carbon dioxide (PaCO2), base excess, peripheral white blood cell (WBC) count, hematocrit, alkaline phosphatase (ALP), γ-glutamyl transpeptidase (γ-GT), lactate dehydrogenase (LDH), C-reactive protein (CRP), serum creatinine, blood urea nitrogen (BUN), calcium, albumin and total bilirubin. Complications included shock, acute respiratory distress syndrome (ARDS; defining as an acute onset of hypoxemia and the ratio of PaO2 to fraction of inspiration oxygen [FiO2] of 90 beats/min, respiratory rate > 20 breaths/min or PaCO2 < 32 mmHg, WBC count < 4.0 × 109/L or >12.0 × 109/L or ≥ 10% bands). Therapies for patients consisted of EN, use of probiotics and prophylactic antibiotics (piperacillin, carbapenem, quinolone and cephalosporin). Definitions Based on the 1992 Summary of the International Symposium on Acute Pancreatitis in Atlanta, (GA, USA), SAP was associated with organ failure and/or local complications such as necrosis, abscess or pseudocyst.1 PaO2 was graded as: low, mean PaO2 < 60 mmHg, or PaO2 < 60 mmHg continuously for more than 48 h; moderate, mean PaO2 60–80 mmHg, or PaO2 < 80 mmHg continuously for less than 48 h; or high, mean PaO2 > 80 mmHg, or PaO2 > 80 mmHg for >48 h. Probiotic application was classified as: early, ≤2 weeks; or late, >2 weeks. Fluid resuscitation was recorded as: early, occurred within 24 h of the admission to achieve central venous pressure >8 cmH2O, urine output > 0.5 mL/kg/h and hematocrit < 44%; or late, not meeting the criteria for early resuscitation. Statistical analysis Statistical analyses were performed using SAS 9.2 (SAS Institute, Cary, NC, USA). Quantitative variables were presented as mean ± standard deviation, and categorical variables were expressed as percentages. Comparisons of quantitative variables between the two groups were conducted using the Mann-Whitney U test, whereas categorical variables were compared with Fisher’s exact test. Statistically significant variables in univariate analysis with a P value of < 0.05 were selected for unconditional multivariate logistic regression analysis to identify the independent risk factors for pancreatic infection in patients with SAP. The optimal cut-off values, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of identified independent predictors of pancreatic infection were determined using the receiver operating characteristic (ROC) curve and the area under the ROC curve was calculated. P ≤ 0.05 was considered statistically significant.
This article is protected by copyright. All rights reserved.
Accepted Article
RESULTS Demographic and clinical characteristics of the patients Finally, a total of 163 patients with SAP were enrolled; among them, 32 (19.6%) with a mean age of 51.9 ± 13.8 years, including 21 men and 11 women, were found to have pancreatic infection. Those in the absence of pancreatic infection included 75 men and 56 women aged 53.3 ± 15.2 years. The demographic and clinical characteristics of the patients with or without pancreatic infection are summarized in Table 1. There were no significant differences in age, gender, etiology of SAP, APACHE II score or Ranson’s score between the two groups (all P > 0.05). However, patients with pancreatic infection had a longer length of hospital stay (43.7 ± 27.4 days vs 22.8 ± 13.0 days, P < 0.0001), higher CTSI (5.59 ± 2.34 vs 3.13 ± 1.04, P < 0.0001) and mortality rate (25.0% vs 3.8%, P = 0.001), respectively, compared with those without pancreatic infection. Laboratory examination results Based on their laboratory examination results, patients with pancreatic infection had lower PaCO2 (31.1 ± 8.9 mmHg vs 34.9 ± 7.1 mmHg, P = 0.037), peripheral WBC count (12.5 × 109/L vs 14.5 × 109/L, P = 0.033) and ALP levels (78.1 ± 60.9 U/L vs 105.7 ± 91.6 U/L) than patients without pancreatic infection, while their LDH (1080.8 ± 864.2 U/L vs 451.1 ± 280.5 U/L) and BUN levels (9.3 ± 6.6 mmol/L vs 7.5 ± 8.8 mmol/L) were much higher (Table 2).
Complications and treatments SIRS, shock and ARF were more commonly seen in patients with pancreatic infection compared with those without pancreatic infection (all P < 0.01, Table 3), while the prevalence of ARDS did not differ between the two groups (P = 0.076). The absence of probiotic application was associated with the lowest probability of pancreatic infection, followed by its early and late application, respectively (21.9% vs 37.5% vs 40.6%). Pancreatic infection also occurred more commonly in patients having low PaO2 than in those with high PaO2, and in patients receiving late fluid resuscitation than in those who received early fluid resuscitation. Administration of EN had no effect on the probability of pancreatic infection (P = 0.626). With regard to antibiotic prophylaxis, the proportion of patients received carbapenem who developed pancreatic infection was the highest among all the patients, while the probability was lowest for those who received quinolone (P = 0.040).
Risk factors associated with pancreatic infection in SAP Using multivariate logistic regression analysis four independent risk factors were identified for pancreatic infection in SAP patients, including LDH levels (OR 1.001, 95% CI 1.000–1.003, P = 0.028), CTSI (OR 1.847, 95% CI 1.157–2.949, P = 0.010), PaO2 (moderate: OR 0.150, 95% CI 0.026–0.852, P = 0.032; and high: OR 0.052, 95% CI 0.007–0.409, P = 0.005) and late fluid resuscitation (OR 7.363, 95% CI 1.654–32.783, P = 0.010) (Table 4). With the following formula: P (pancreatic infection︱ψ) = exp(－3.8456 － 0.00148 × LDH [U/L] + 0.6137 × CTSI + 1.9964 × I[fluid resuscitation = late] － 1.8965 × I[PaO2 = moderate] －2.9527 × I[PaO2 = high])___________________________ 1 + exp(－3.8456 － 0.00148 × LDH [U/L] + 0.6137 × CTSI + 1.9964 × I[fluid resuscitation = late] － 1.8965 × I[PaO2 = moderate] －2.9527 × I[PaO2 = high])
This article is protected by copyright. All rights reserved.
Accepted Article
I[fluid resuscitation = late] or I[PaO2 = moderate] or I[PaO2 = high]) was regarded as 1 there was a late resuscitation or moderate or high PaO2, respectively; otherwise, the value was 0. With a cut-off value of 0.393, the sensitivity, specificity, PPV and NPV in predicting pancreatic infection in patients with SAP were 84%, 97%, 88%, and 96%, respectively, with the AUROC of 0.923 (95% CI 0.848–0.998). DISCUSSION Pancreatic infection could be observed in up to 30% of the patients with necrotizing pancreatitis.5 Pancreatic infection itself and its septic complications are the most important causes of death in SAP, leading to a mortality rate of 60–80% in patients with AP.4,20–23 Its development greatly influences the prognosis of and determines the therapeutic approaches for the patients. Therefore, it is critically important to identify noninvasive and universally available biochemical parameters as well as scoring systems that can predict the development of secondary pancreatic infection in these patients. Currently, few biochemical parameters such as CRP, interleukin (IL)-6, nitric oxide (NO) and procalcitonin, have been investigated for their clinical relevance in diagnosis of secondary pancreatic infection.24–27 In the present study, we focused on the efficacy of several parameters for predicting pancreatic infection in patients with SAP, and found that an increased LDH, high CTSI, hypoxemia and delayed fluid resuscitation were independent risk factors for pancreatic infection. The extent of pancreatic necrosis is correlated with the frequency of organ failure, which is approximately 20% when pancreatic necrosis is
Article Type: Original Article Risk factors for pancreatic infection in patients with severe acute pancreatitis: An analysis of 163 cases1 Yan Bo ZENG,* Xian Bao ZHAN,* Xiao Rong GUO, Hua Gao ZHANG, Yan CHEN, Quan Cai CAI, Zhao Shen LI Department of Gastroenterology, Changhai Hospital, Second Military Medical University, Shanghai, China *These two authors contributed equally to this work. Correspondence to: Zhao Shen LI, Department of Gastroenterology, Changhai Hospital, Second Military Medical University, 168 Changhai Road, Shanghai 200433, China. Email: [email protected] Conflicts of interest: None. Running title: Risk factors for pancreatic infection
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/1751-2980.12150
This article is protected by copyright. All rights reserved.
Accepted Article
ABSTRACT OBJECTIVE: We aimed to identify the risk factors for predicting pancreatic infection in patients with severe acute pancreatitis (SAP). METHODS: One hundred and sixty-three patients with SAP were included and divided into two groups based on the presence or absence of pancreatic infection. Demographic and clinical characteristics, laboratory examination results, complications and treatment modalities of these patients were collected from their medical records. Variables were initially screened by univariate analysis and those with statistical significance were then filtered by multivariate analysis to determine the independent risk factors for pancreatic infection in SAP. RESULTS: Patients having SAP with pancreatic infection had lower partial pressure of arterial CO2, peripheral white blood cell count and alkaline phosphatase levels, as well as a higher computed tomography severity index (CTSI) than those without pancreatic infection, while their lactate dehydrogenase levels and partial pressure of arterial O2 (PaO2) were much higher. Pancreatic infection also occurred more commonly in patients receiving late fluid resuscitation than in those who received early fluid resuscitation.Multivariate analyses revealed that increased lactate dehydrogenase level, high CTSI, delayed fluid resuscitation and low PaO2 were independent risk factors for pancreatic infection in SAP. The sensitivity, specificity, positive and negative predictive values for a model combining the parameters in predicting pancreatic infection were 84%, 97%, 88%, and 96%, respectively, with a cut-off value of 0.393, and the AUROC was 0.923. CONCLUSION: Increased lactate dehydrogenase, high CTSI, delayed fluid resuscitation and hypoxemia are independent risk factors for predicting pancreatic infection in patients with SAP. Keywords: acute necrotizing pancreatitis, diagnosis, hypoxemia, pancreatitis, pancreatic infection, risk factors
This article is protected by copyright. All rights reserved.
Accepted Article
INTRODUCTION A sudden inflammation of the pancreas, known as acute pancreatitis (AP), may variably affect adjacent tissues and even remote organs.1 Severe acute pancreatitis (SAP) has been reported to occur in less than 30% of the AP cases, however, it is associated with a high mortality rate.2,3 SAP usually manifests as local complications such as pancreatic necrosis, pseudocyst or abscess, and/or advanced organ failure including renal, cardiocirculatory and pulmonary dysfuction as well as sepsis, etc.4,5 Multiple organ failure (MOF) and severe infectious complications are life-threatening and often associated with pancreatic necrosis. Infected pancreatic necrosis has been found in 30–70% of the patients with necrotizing pancreatitis, and it significantly increases the incidence of systematic complications and is a leading cause for a high mortality in SAP.5–7 Prospective studies8–12 have identified a number of clinical prognostic scores and serum markers that can be used for assessing and monitoring AP. However, the potential of these parameters to identify patients at risk of developing infected pancreatic necrosis remains unknown. Over the past decades therapeutic strategies for SAP have been developed, and conservative therapies now play a prominent role in treating SAP. However, it is still controversial over the use of some modalities such as antibiotic prophylaxis, probiotics and enteral nutrition (EN) for the prevention of pancreatic infection and a decrease of mortality in AP patients.13–17 Diagnosis of pancreatic infection is crucial in determining the therapeutic approaches for SAP. Early diagnosis of and proper therapies for infected pancreatic necrosis are important for improving the prognosis and decreasing the morbidity in such patients. Serum markers that facilitate early diagnosis of pancreatic infection would be clinically useful. In this retrospective study, we aimed to assess the predictive value of a variety of parameters including biochemical markers, scoring systems, complications and treatments for pancreatic infection in patients with SAP treated with conservative therapies, and to identify the independent risk factors for pancreatic infection.
PATIENTS AND METHODS Patients From October 2008 to December 2011, 252 consecutive patients were admitted to Changhai Hospital, Second Military Medical University (Shanghai, China) due to SAP. Of these patients, 163 (96 men and 67 women) aged ≥18 years who had not been treated with probiotics or prophylactic antibiotics before their admission were finally recruited in the study. All these 163 patients were further divided into two groups, that is, pancreatic infection group (n = 32) and non-pancreatic infection group (n = 131). When pancreatic infection was suspected by persisting or new-onset clinical manifestations and/or laboratory results of sepsis after ruled out infection from other sources, pancreatic infection was confirmed by positive results from bacteria culture obtained from computed tomography (CT)-guided or endoscopic ultrasound-guided fine needle aspiration culture. The study was approved by the Institutional Ethics Committee of Changhai Hospital, Second Military Medical University. Data collection Demographic and clinical characteristics of the patients, their laboratory results, complications and treatment modalities were collected by reviewing the medical records of the patients. Clinical data consisted of etiology of SAP, length of hospital stay, mortality, the Acute Physiology and Chronic Health Evaluation (APACHE) II score,18 Ranson’s score9 and CT severity index (CTSI)19. The APACHE II score and the Ranson’s score were determined within This article is protected by copyright. All rights reserved.
Accepted Article
the first 24 h and 48 h after admission, respectively. Contrast-enhanced CT was performed in all the patients and CTSI was determined within 48 h after their admission to assess the extent of pancreatic inflammation and necrosis. Laboratory data were obtained within the initial 24 h from the onset of disease, including pH value, partial pressure of arterial oxygen (PaO2), arterial partial pressure of carbon dioxide (PaCO2), base excess, peripheral white blood cell (WBC) count, hematocrit, alkaline phosphatase (ALP), γ-glutamyl transpeptidase (γ-GT), lactate dehydrogenase (LDH), C-reactive protein (CRP), serum creatinine, blood urea nitrogen (BUN), calcium, albumin and total bilirubin. Complications included shock, acute respiratory distress syndrome (ARDS; defining as an acute onset of hypoxemia and the ratio of PaO2 to fraction of inspiration oxygen [FiO2] of 90 beats/min, respiratory rate > 20 breaths/min or PaCO2 < 32 mmHg, WBC count < 4.0 × 109/L or >12.0 × 109/L or ≥ 10% bands). Therapies for patients consisted of EN, use of probiotics and prophylactic antibiotics (piperacillin, carbapenem, quinolone and cephalosporin). Definitions Based on the 1992 Summary of the International Symposium on Acute Pancreatitis in Atlanta, (GA, USA), SAP was associated with organ failure and/or local complications such as necrosis, abscess or pseudocyst.1 PaO2 was graded as: low, mean PaO2 < 60 mmHg, or PaO2 < 60 mmHg continuously for more than 48 h; moderate, mean PaO2 60–80 mmHg, or PaO2 < 80 mmHg continuously for less than 48 h; or high, mean PaO2 > 80 mmHg, or PaO2 > 80 mmHg for >48 h. Probiotic application was classified as: early, ≤2 weeks; or late, >2 weeks. Fluid resuscitation was recorded as: early, occurred within 24 h of the admission to achieve central venous pressure >8 cmH2O, urine output > 0.5 mL/kg/h and hematocrit < 44%; or late, not meeting the criteria for early resuscitation. Statistical analysis Statistical analyses were performed using SAS 9.2 (SAS Institute, Cary, NC, USA). Quantitative variables were presented as mean ± standard deviation, and categorical variables were expressed as percentages. Comparisons of quantitative variables between the two groups were conducted using the Mann-Whitney U test, whereas categorical variables were compared with Fisher’s exact test. Statistically significant variables in univariate analysis with a P value of < 0.05 were selected for unconditional multivariate logistic regression analysis to identify the independent risk factors for pancreatic infection in patients with SAP. The optimal cut-off values, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of identified independent predictors of pancreatic infection were determined using the receiver operating characteristic (ROC) curve and the area under the ROC curve was calculated. P ≤ 0.05 was considered statistically significant.
This article is protected by copyright. All rights reserved.
Accepted Article
RESULTS Demographic and clinical characteristics of the patients Finally, a total of 163 patients with SAP were enrolled; among them, 32 (19.6%) with a mean age of 51.9 ± 13.8 years, including 21 men and 11 women, were found to have pancreatic infection. Those in the absence of pancreatic infection included 75 men and 56 women aged 53.3 ± 15.2 years. The demographic and clinical characteristics of the patients with or without pancreatic infection are summarized in Table 1. There were no significant differences in age, gender, etiology of SAP, APACHE II score or Ranson’s score between the two groups (all P > 0.05). However, patients with pancreatic infection had a longer length of hospital stay (43.7 ± 27.4 days vs 22.8 ± 13.0 days, P < 0.0001), higher CTSI (5.59 ± 2.34 vs 3.13 ± 1.04, P < 0.0001) and mortality rate (25.0% vs 3.8%, P = 0.001), respectively, compared with those without pancreatic infection. Laboratory examination results Based on their laboratory examination results, patients with pancreatic infection had lower PaCO2 (31.1 ± 8.9 mmHg vs 34.9 ± 7.1 mmHg, P = 0.037), peripheral WBC count (12.5 × 109/L vs 14.5 × 109/L, P = 0.033) and ALP levels (78.1 ± 60.9 U/L vs 105.7 ± 91.6 U/L) than patients without pancreatic infection, while their LDH (1080.8 ± 864.2 U/L vs 451.1 ± 280.5 U/L) and BUN levels (9.3 ± 6.6 mmol/L vs 7.5 ± 8.8 mmol/L) were much higher (Table 2).
Complications and treatments SIRS, shock and ARF were more commonly seen in patients with pancreatic infection compared with those without pancreatic infection (all P < 0.01, Table 3), while the prevalence of ARDS did not differ between the two groups (P = 0.076). The absence of probiotic application was associated with the lowest probability of pancreatic infection, followed by its early and late application, respectively (21.9% vs 37.5% vs 40.6%). Pancreatic infection also occurred more commonly in patients having low PaO2 than in those with high PaO2, and in patients receiving late fluid resuscitation than in those who received early fluid resuscitation. Administration of EN had no effect on the probability of pancreatic infection (P = 0.626). With regard to antibiotic prophylaxis, the proportion of patients received carbapenem who developed pancreatic infection was the highest among all the patients, while the probability was lowest for those who received quinolone (P = 0.040).
Risk factors associated with pancreatic infection in SAP Using multivariate logistic regression analysis four independent risk factors were identified for pancreatic infection in SAP patients, including LDH levels (OR 1.001, 95% CI 1.000–1.003, P = 0.028), CTSI (OR 1.847, 95% CI 1.157–2.949, P = 0.010), PaO2 (moderate: OR 0.150, 95% CI 0.026–0.852, P = 0.032; and high: OR 0.052, 95% CI 0.007–0.409, P = 0.005) and late fluid resuscitation (OR 7.363, 95% CI 1.654–32.783, P = 0.010) (Table 4). With the following formula: P (pancreatic infection︱ψ) = exp(－3.8456 － 0.00148 × LDH [U/L] + 0.6137 × CTSI + 1.9964 × I[fluid resuscitation = late] － 1.8965 × I[PaO2 = moderate] －2.9527 × I[PaO2 = high])___________________________ 1 + exp(－3.8456 － 0.00148 × LDH [U/L] + 0.6137 × CTSI + 1.9964 × I[fluid resuscitation = late] － 1.8965 × I[PaO2 = moderate] －2.9527 × I[PaO2 = high])
This article is protected by copyright. All rights reserved.
Accepted Article
I[fluid resuscitation = late] or I[PaO2 = moderate] or I[PaO2 = high]) was regarded as 1 there was a late resuscitation or moderate or high PaO2, respectively; otherwise, the value was 0. With a cut-off value of 0.393, the sensitivity, specificity, PPV and NPV in predicting pancreatic infection in patients with SAP were 84%, 97%, 88%, and 96%, respectively, with the AUROC of 0.923 (95% CI 0.848–0.998). DISCUSSION Pancreatic infection could be observed in up to 30% of the patients with necrotizing pancreatitis.5 Pancreatic infection itself and its septic complications are the most important causes of death in SAP, leading to a mortality rate of 60–80% in patients with AP.4,20–23 Its development greatly influences the prognosis of and determines the therapeutic approaches for the patients. Therefore, it is critically important to identify noninvasive and universally available biochemical parameters as well as scoring systems that can predict the development of secondary pancreatic infection in these patients. Currently, few biochemical parameters such as CRP, interleukin (IL)-6, nitric oxide (NO) and procalcitonin, have been investigated for their clinical relevance in diagnosis of secondary pancreatic infection.24–27 In the present study, we focused on the efficacy of several parameters for predicting pancreatic infection in patients with SAP, and found that an increased LDH, high CTSI, hypoxemia and delayed fluid resuscitation were independent risk factors for pancreatic infection. The extent of pancreatic necrosis is correlated with the frequency of organ failure, which is approximately 20% when pancreatic necrosis is
