Acute pancreatitis - predicting the severity of the disease.

RATIONALE

Acute pancreatitis (AP) is a serious acute abdominal disease. AP is often referred to as an unpredictable illness, which can take a mild to severe (fatal) course.

AIMS OF THE STUDY

1. To identify clinical parameters that are significantly related to the clinical course of acute pancreatitis. 2) To compile a scoring system enabling the severity of AP to be predicted when the patient is first admitted to hospital.

METHODS

Analysis of available publications and clinical guidance, and retrospective analysis of data on patients hospitalised with AP at our clinic enable us to identify clinical details and laboratory results recorded at the time of patients' admission to hospital that are related to the subsequent severity of the disease. For the purposes of statistical analysis, the sample of patients was divided into two groups: group A (mild AP, without local or organ complications), group B (moderately severe and severe AP with local and/or organ complications).

PATIENT GROUPS AND RESULTS

In total, between 01.01.2013 and 30.06.2022, 312 patients with acute pancreatitis were allocated to the retrospective-prospective study sample. 74 % (231/312) of these patients were allocated to group A and 26 % (81/312) were allocated to group B. Univariate analysis of the data collected on the patient sample identified 5 parameters that are statistically significantly associated with the severity of the clinical course of the disease. Presence of SIRS on admission (A vs B, Odds ratio 10.787, 95% CI 5.09-22.85, p < 0.0001), diabetes mellitus type 2 in case history (A vs B, Odds ratio 7.703, 95% CI 3.04-19.51, p 2 mmol/l (A vs B, Odds ratio 3.293, 95% CI 1.59-6.82, p = 0.0013).In order to develop a scoring system, each of these parameters was allocated a points value based on its Odds ratio (OR): presence of SIRS 3 points, hypocalcaemia 3 points, diabetes mellitus type 2 in case history 2 points, urea concentration > 8 mmol/l 1 point and lactate concentration > 2 mmol/l 1 point. The authors refer to their scoring system as The Acute Pancreatitis Admission Score (APAS). The accuracy of APAS was modelled for various cut off values. Across the whole sample, we ascertained that an APAS ≥ 4 points predicts moderately severe or severe AP with a sensitivity of 81 % (95% CI: 71 - 89 %) and specificity of 87 % (95 CI: 81 - 91 %). The positive predictive value (PPV) of APAS ≥ 4 is 0.68, while its negative predictive value (NPV) is 0.93 and accuracy 0.85 (95% CI 0.81 - 0.89).

CONCLUSION

In this study we identify significant simple clinical and laboratory parameters that are commonly tested as part of an initial examination when admitting a patient with AP to hospital. Having identified these parameters we are able to establish a simple scoring system that is able to predict the severity of the course of AP at the moment of hospitalisation (Tab. 5, Fig. 2, Ref. 27).