We included 424 patients, with biliary pancreatitis seen during 2018 and 2019; as reported in the literature, with biliary etiology as the main cause of acute pancreatitis. The mean age of the patients was 60.69 years, in relation to previous studies that have shown that pancreatitis affects people of productive age (16). As in other studies, a tendency was found to affect females more (62.74%), possibly because women have a higher incidence of benign biliary disease (17).
The mortality reported in the literature for pancreatitis may be 1% in general (6), although directly related to severity, in its serious forms it may reach up to 30% (18); in our case the mortality in the group of patients studied was 5.8%; this is in relation to the fact that cases of severe pancreatitis according to the Atlanta classification were higher in this population and that our center is a regional reference center.
We found that 61% of the pancreatitis was severe, that is, 258 patients had persistent organ failure; of these, 87% of the patients were classified in this group, because they had ventilatory of failure (less than 300 according to PaFi). This result is notorious, as the data found in our population show a much higher percentage of pancreatitis classified as severe, when compared to other studies (2,19); the reason for these findings is related to lower arterial oxygen partial pressure when compared other populations due to the altitude above sea level, so there is an overestimation of severity.
When we assess the predictability of the three scores to predict adverse outcomes, we find that Marshall’s performance is below expectations and below what has been observed in other studies (22). The same may be related to decreased PaFi because when the PaFi is below 300 it should be assigned 2 points in the Marshall score, this implies that this patient is already classified between moderately severe and if these values persist, as severe (≥2 defines the presence of organic failure). Morevoer, APACHE II has a better performance in the prediction of unfavorable outcomes, even though the PaFi relationship is also within its variables, this is, because it is not the only variable that must be altered so that the score is greater than or equal to 8 which is the score that defines pancreatitis as severe, but other parameters must be altered. In the case of BISAP we observed that it is the best performing score to predict adverse outcomes in our population with an AUC 0.744, probably because it does not take PaFi into account within its variables. So, in our midst it is an alternative to consider.
When assessing whether there were differences between AUC the three scores to predict unfavorable outcomes, we can show that there is a statistically significant higher performance of BISAP over Marshall, and there are no statistically significant differences between BISAP and APACHE II.
Overestimation of the severity of acute pancreatitis leads to higher hospital costs and leads these patients to occupy beds in the intensive care unit unnecessarily, thus increasing the days of hospital stays, exposing patients to complications associated with hospitalization such as infections and thrombotic diseases.
The limitations of our study lie in its retrospective nature and that we only included patients with biliary pancreatitis, which are predominant in our context, pancreatitis with other kinds of etiology should be evaluated in another study.
These findings, allow us to suggest that the BISAP score be used for stratification of severity in all patients with acute biliary pancreatitis in populations where PaFi may be altered due to decreased partial pressure of oxygen while other studies are being conducted evaluating others scores with the PaFi adjusted to the partial pressure of oxygen according to atmospheric pressure. In addition, BISAP is easy to establish and has been evaluated for the prediction of adverse outcomes being an alternative comparable to APACHE II and superior to other scores (19–21).