Limits...
Radial EUS Examination Can be Helpful in Predicting the Severity of Acute Biliary Pancreatitis

View Article: PubMed Central - PubMed

ABSTRACT

We investigated the utility of noncontrast enhanced endosonography (EUS) in predicting the severity of acute pancreatitis (AP) during the first 72 to 96 h of admission.

In total, 187 patients with acute biliary pancreatitis were included. The patients were classified into 2 groups as having severe and mild AP according to the Modified Glasgow scoring and computerized tomography severity index (SI). The 158 cases with mild and 29 cases with severe AP had a similar age and sex distribution.

Although none of the cases with mild AP developed morbidity and death, of the cases with severe AP, 16 developed serious morbidities and 5 died. On EUS examination, we looked for parenchymal findings, peripancreatic inflammatory signs, free or loculated fluid collections, and abnormalities of the common bile duct and the pancreatic channel. Statistical analysis indicated a significant relationship between the severity of AP with diffuse parenchymal edema, periparenchymal plastering, and/or diffuse retroperitoneal free fluid accumulation, and peri-pancreatic edema. We also defined an EUSSI and found that the EUSSI had sensitivity of 89.7%, specificity of 84.2%, positive predictivity value (PPV) of 88.9%, negative predictivity value (NPV) of 91.2%, and an accuracy of 87.9% in the differentiation of mild and severe AP. We found that the EUSSI had an accuracy of 72.4%, sensitivity of 75.4%, specificity of 65.1%, PPV of 69.3%, and NPV of 73.1% for determining mortality.

Our data suggest that EUS allowed us to accurately predict the severity and mortality in nearly 90% of cases with AP.

No MeSH data available.


Related in: MedlinePlus

(A) Pancreatic necrosis (>30 mm) is shown. (B) CT appearance of the same patient indicating huge amount of parenchymal necrosis in the pancreas. CT = computerized tomography.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4998239&req=5

Figure 8: (A) Pancreatic necrosis (>30 mm) is shown. (B) CT appearance of the same patient indicating huge amount of parenchymal necrosis in the pancreas. CT = computerized tomography.

Mentions: We enrolled 187 patients with a diagnosis of acute biliary pancreatitis. We classified our cases as SAP if the Balthazar CTSI indicated a value of >6 and/or modified Glasgow scoring showed >3 criteria. According to this classification, 29 patients (15.5%) had SAP, and 158 (84.5%) had MAP. There was no difference with regard to the age and sex distribution of the patients with mild and SAP. The mean CRP levels were significantly higher in the cases with SAP (34 ± 16 mg/dL) than in the cases with MAP (7 ± 6 mg/dL) (P < 0.05). Morever, the CRP levels correlated well with the EUSSI and CTSI (r:0.38, P:0.002). Twenty-five patients out of the 29 (85.2%) with SAP according to Glasgow scoring had a Balthazar CTSI ≥7. We used Marshall scoring to assess organ dysfunction in our cases at 48 to 72 h after admission to our clinics, which indicated a significant difference between the patients with MAP and SAP (0.5 ± 0.1 vs 3.1 ± 1.3, respectively, P < 0.05). Eight patients with SAP progressed to cardiac failure. Ten cases developed renal failure persisting for >48 h. Eleven patients with SAP developed several morbidities needing several surgical and or percutaneous procedures. Five of these patients died. No patient with MAP died or developed serious morbidities. Comparing patients with mild and SAP, there were also significant statistical differences regarding Marshall scores, morbidity and mortality figures (Figures 1–8), and the EUSSI and CTSl. Moreover, the mean hospitalization period was 17 ± 6 days (range: 49–12 days) in patients with SAP and 5 ± 1.8 days in patients with MAP (range: 4–8 days) (P < 0.001) (Table 3).


Radial EUS Examination Can be Helpful in Predicting the Severity of Acute Biliary Pancreatitis
(A) Pancreatic necrosis (>30 mm) is shown. (B) CT appearance of the same patient indicating huge amount of parenchymal necrosis in the pancreas. CT = computerized tomography.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4998239&req=5

Figure 8: (A) Pancreatic necrosis (>30 mm) is shown. (B) CT appearance of the same patient indicating huge amount of parenchymal necrosis in the pancreas. CT = computerized tomography.
Mentions: We enrolled 187 patients with a diagnosis of acute biliary pancreatitis. We classified our cases as SAP if the Balthazar CTSI indicated a value of >6 and/or modified Glasgow scoring showed >3 criteria. According to this classification, 29 patients (15.5%) had SAP, and 158 (84.5%) had MAP. There was no difference with regard to the age and sex distribution of the patients with mild and SAP. The mean CRP levels were significantly higher in the cases with SAP (34 ± 16 mg/dL) than in the cases with MAP (7 ± 6 mg/dL) (P < 0.05). Morever, the CRP levels correlated well with the EUSSI and CTSI (r:0.38, P:0.002). Twenty-five patients out of the 29 (85.2%) with SAP according to Glasgow scoring had a Balthazar CTSI ≥7. We used Marshall scoring to assess organ dysfunction in our cases at 48 to 72 h after admission to our clinics, which indicated a significant difference between the patients with MAP and SAP (0.5 ± 0.1 vs 3.1 ± 1.3, respectively, P < 0.05). Eight patients with SAP progressed to cardiac failure. Ten cases developed renal failure persisting for >48 h. Eleven patients with SAP developed several morbidities needing several surgical and or percutaneous procedures. Five of these patients died. No patient with MAP died or developed serious morbidities. Comparing patients with mild and SAP, there were also significant statistical differences regarding Marshall scores, morbidity and mortality figures (Figures 1–8), and the EUSSI and CTSl. Moreover, the mean hospitalization period was 17 ± 6 days (range: 49–12 days) in patients with SAP and 5 ± 1.8 days in patients with MAP (range: 4–8 days) (P < 0.001) (Table 3).

View Article: PubMed Central - PubMed

ABSTRACT

We investigated the utility of noncontrast enhanced endosonography (EUS) in predicting the severity of acute pancreatitis (AP) during the first 72 to 96&#8202;h of admission.

In total, 187 patients with acute biliary pancreatitis were included. The patients were classified into 2 groups as having severe and mild AP according to the Modified Glasgow scoring and computerized tomography severity index (SI). The 158 cases with mild and 29 cases with severe AP had a similar age and sex distribution.

Although none of the cases with mild AP developed morbidity and death, of the cases with severe AP, 16 developed serious morbidities and 5 died. On EUS examination, we looked for parenchymal findings, peripancreatic inflammatory signs, free or loculated fluid collections, and abnormalities of the common bile duct and the pancreatic channel. Statistical analysis indicated a significant relationship between the severity of AP with diffuse parenchymal edema, periparenchymal plastering, and/or diffuse retroperitoneal free fluid accumulation, and peri-pancreatic edema. We also defined an EUSSI and found that the EUSSI had sensitivity of 89.7%, specificity of 84.2%, positive predictivity value (PPV) of 88.9%, negative predictivity value (NPV) of 91.2%, and an accuracy of 87.9% in the differentiation of mild and severe AP. We found that the EUSSI had an accuracy of 72.4%, sensitivity of 75.4%, specificity of 65.1%, PPV of 69.3%, and NPV of 73.1% for determining mortality.

Our data suggest that EUS allowed us to accurately predict the severity and mortality in nearly 90% of cases with AP.

No MeSH data available.


Related in: MedlinePlus