The influence of treatment access regulation and technological resources on the mortality profile of acute biliary pancreatitis

Purpose: The influence of treatment access regulation and technological resources on the mortality profile of acute biliary pancreatitis (ABP) was evaluated. Methods: The cases seen in a tertiary hospital were studied during two periods of time: 1995-1999 and 2000-2004, i.e., before and after the implementation of medical regulation. Results: Among the 727 patients with acute pancreatitis, 267 had ABP and were classified according to APACHE II scores. The cases being referred to the tertiary hospital decreased from 441 to 286 (p < 0.001). The patients’ profile regarding age, gender, severity, cholestasis incidence and mortality were similar during the first and second periods of study (n = 154 and n = 113, respectively). The number of patients with hematocrit > 44% was smaller during the second study period (p<0.002). The use of magnetic resonance cholangiography, videolaparoscopic cholecystectomy, and access to the ICU were found to be more frequent during the second study period. Regarding the deaths occurring within 14 days of hospitalisation, 73.4% and 81.3% were observed during the first and second study periods, respectively. Conclusion: Since the improvement in clinical and technological approach was not enough to modify the mortality profile of ABP, further studies on the treatment of inflammatory responses should be carried out.


Introduction
In the great majority of cases, severe acute pancreatitis occurs in two phases.During the first couple of weeks, a systemic inflammatory reaction occurs due to the release of various pro-inflammatory cytokines, such as tumour necrosis factor (TNF), interleukins (IL-1, IL-2, IL-6), nitric oxide, oxygen-reactive radicals, and arachidonic acid metabolites. 1. The overall mortality of acute pancreatitis is 5% (ranging from 2 to 9%), including all other aetiologies. 3The mortality occurring during the early phase (two first weeks) ranges from 0 to 82.3%. 4,5][7] In the literature there is no study on the mortality profile of acute biliary pancreatitis (ABP) regarding early and late phases, which justifies an evaluation of the possible determinants involved.
In 2000, in an attempt to improve the diagnostic and treatment conditions of acute severe cases seen in the emergency settings, a medical regulation was applied to the ambulance service emergency care (SAMU) which covers the greater Ribeirão Preto. 8edical regulation has the objective of coordinating the available resources according to the health problem involved, so that the best diagnostic and therapeutic alternatives can be provided.Since the implementation of the medical regulation, the Emergency Unit of the Ribeirão Preto University Hospital has been considered a reference point in terms of high complexity cases.As a result, the facilities were architecturally and technologically improved to provide clinical stabilisation and intensive care. 8,9oncomitantly, this process of organising the health care system and establishing inpatient references may benefit patients with biliary lithiasis because of the extensive access to videolaparoscopic cholecystectomy.Outpatient care and long-range care have contributed to reducing the waiting list for cholecystectomy among these low-risk surgical patients. 10In addition, magnetic resonance cholangiography was gradually implemented as part of examination procedures for biliary duct investigation in the cases of extra-hepatic cholestasis, thus reducing the need for more invasive procedures. 11,12ithin this context, it is expected that not only the complications resulting from biliary lithiasis will be reduced, particularly acute pancreatitis, but also that the access to emergency units will become more rapid.
The objective of the present study was to evaluate the influence of medical regulation, hospital organisation, and resources for cholelithiasis treatment on the mortality profile of ABP during early and later phases.

Methods
The medical records of patients over 12 years old who were admitted to the Ribeirão Preto University Hospital or Emergency Unit between January 1 st , 1995 and December 31 st , 2004 for ABP treatment were retrospectively reviewed.Information was collected according to the International Disease Code (ICD 9 -577.0 and ICD 10 -K 85.0).
The study was divided into two periods of time: 1995 to 1999 and 2000 to 2005.This temporal division was based on the date of implementation of SAMU and medical regulation in the greater Ribeirão Preto area, that is, the periods before and after the year 1999.
The individual data were gathered according to the medical record review protocol.Epidemiological and clinical information, image and pathological exams, and diagnostic and therapeutic procedures were tabulated on electronic spreadsheets (Microsoft Excel, Seattle, USA).
ABP diagnosis was based on the presence of at least two or three criteria: 1) upper abdominal pain, 2) serum levels of amylase and/or lipase > 3 times their normal values; 3) computed tomography (CT) or magnetic resonance (MR) findings commonly observed in acute pancreatitis. 3The presence of biliary lithiasis was diagnosed using abdominal CT, MR or ultrasonography.ABP severity was evaluated according to both APACHE II score 13 and Atlanta classification. 14Data for the first 24 hours of hospitalisation were collected for evaluation using APACHE II score.If this information was not available, a corresponding normal value was used instead.
Mortality was considered precocious if death occurred within 14 days following admission.The APACHE II score was calculated using an electronic plan available at www.sfar.org/scores2.
The differences between the mean APACHE II values and the percent death risk were determined by the Student "t" test, the frequencies were evaluated by the Fisher exact test and the Pearson Chi-square test, and the medians were evaluated by the Mann-Whitney test.
The values obtained were expressed as mean and standard deviation, with the level of significance set at α = 0.05 (P < 0.05).The frequencies were expressed as percent values.Bonferroni correction was applied to multiple intraperiod comparisons.A receiver operating characteristic (ROC) curve was constructed to validate the APACHE I score.
With respect to the image exams used to diagnose biliary involvement in the cases of acute pancreatitis, a significant increase was observed in magnetic resonance cholangiography.Among the therapeutic procedures, videolaparoscopic cholecystectomy was increasingly used while both open cholecystectomy and endoscopic retrograde cholangiopancreatography were decreasingly used following a 72-hour period (Table 1).
Cholestasis was observed in almost 60% of the patients during both study periods.Investigation by means of MRC was significantly more frequent during the second study period, whereas a non-significant decrease in the use of ERCP and intra-operative cholangiography was observed.The incidence of choledocholithiasis was low during both study periods, with no significant difference between them (Table 2).
The use of intensive care for patients with severe ABP increased significantly during the second study period, although only 40% of the cases had been successful.Despite being statistically equivalent, the estimated 3-day period on a waiting list was found to be long, even during the second study period (Table 3).
The overall mortality rate for ABP cases was 11.6%.Mild ABP (APACHE II < 8) accounted for 68.2% (mortality rate of 3.8%) and 67.3% (mortality rate of 3.9%) of all cases, respectively, during the first and second study periods.In contrast, severe ABP represented 31.8%(mortality rate of 22.4%) and 32.7% (mortality rate of 35.1%) of all cases, respectively, during the first and second study periods.No significant difference was found between study periods (Figure 2).
Approximately 70% of the deaths occurred within two weeks after the onset of the disease.During this first phase, 69.5% of the deaths were related to the systemic inflammatory response syndrome (SIRS), whereas sepsis was predominant (87.5%) during the second phase (Table 4).
A ROC curve was drawn for the APACHE II score.The sensitivity (74.2%) and specificity (81.2%) for the prediction of death were based on the APACHE II score (> 8.5) (Figure 3).

Discussion
The evaluation of the influence of organization of the health system, of hospital service and of its technological incorporation into the profile of severity and the mortality rate in ABP is a novelty.When the present study was formulated the expectation was that the incidence of ABP cases would be reduced among patients of good anaesthetic and surgical risk as a function of the expanded access to the treatment of colelithiasis by videolaparoscopy on an ambulatory basis and eventually due to the increased frequency of more serious cases at the reference hospital because of the application of medical regulation 8,9,10 .
The objective of this two-phased study was to determine the impact of a more extensive access to the treatment of symptomatic cholelithiasis on the mortality profile and rates of ABP.This includes videolaparoscopic cholecystectomy and non-invasive evaluation of the biliary ducts by means of MRC.By comparing the two study periods, it can be seen that the number of cholecystectomies was increased in the Ribeirão Preto region.Based on the information obtained from the Data Processing Centre of the Department of Social Medicine, School of Medicine of Ribeirão Preto, University of São Paulo, cholecystectomy was performed 12,936 times between 1995 and 1999 and 15,327 times between 2000 and 2004, an increase of 18.5%.The access to videolaparoscopic cholecystectomy increased from 29% to 64% at the Ribeirão Preto University Hospital.
The access to health care services by means of medical regulation has significantly improved the working conditions as well as the emergency care unit.Between the years of 1999 and 2000, a significant reduction in the number of consultations was observed, from 114,311 to 56,752.Also, the number of inpatient beds decreased from 113.3 to 97.1%.Concomitantly, the hospital wards were redefined and the number of beds for critical patients was increased from 16 (7.1%)to 43 (21%) units compared to the total inpatient beds. 8,9ithin this changing scenario, an overall reduction of 35.1% was observed in all cases of acute pancreatitis referred to the emergency care unit after implementation of both medical regulation and SAMU, despite the fact that the general population increased by 11.5% during the study period.In addition, alcohol-related acute pancreatitis and biliary pancreatitis respectively decreased by 55.3% and 26.6%.However, ABP was found to be slightly increased (4.6%) during the 2000-2004 study period.
Because of the decreased number of patients, the more severe cases of ABP were expected to be referred to the emergency care unit.However, this fact was not observed.Not only the proportion of cases of mild ABP seen during both the first and second periods of study, but also the disease profile regarding age, gender, cholestasis incidence, severity, APACHE II score, and mortality were the same.However, a lower number of patients with haemoconcentration (P < 0.002) was seen in the second study period, thus indicating improvement in the clinical care services.
Even though clinical care was improved and access to MRC, VLC, and intensive care was significantly extended, the mortality profile of ABP was not sufficiently modified in the second-period cases despite the resources available.
During both study periods, the overall mortality rate for ABP was 11.6%, which is slightly higher than the maximum value reported in the literature on acute pancreatitis (ranging from 2% to 9%). 3 No difference in the mortality rate of patients with mild ABP was found between the two study periods, a finding also observed in severe ABP.The overall mortality rate for patients with mild ABP was 3.8%, similar to that reported elsewhere (3.0%, ranging from 1% to 7%). 3 In the present study, the overall mortality rate for severe ABP was 27.9%, which corresponds to the limits found in the literature (17%, ranging from 8% to 39%). 3 However, significant differences in age and APACHE II score were found between patients who survived and those who died, a finding already observed in other study on pancreatitis of various aetiologies. 2he determinants of high mortality during the early phase of ABP are not known.Clinical stabilisation can be an important factor to avoid multiple organ failure (MOF), thus reducing the mortality rate. 5Another factor playing a relevant role in ABP evolution is the time elapsed between onset of the symptoms and hospitalisation, including the intensive care. 2 Although the access to intensive care was found to be extensive during the second study period, the time elapsed between admission to the hospital and referral to intensive care did not change, with the first 72 hours remaining uncovered and corresponding to the period during which 33.3% and 50.0% of the deaths were concentrated in the first and second phase of the study, respectively.Of the patients who died, 70% were males over 70 years old (mean age: 74 + 10.4 years).
There is a great discrepancy between the mortality and morbidity profiles of acute severe pancreatitis.Mortality in the early phase ranged from 0 to more than 80.0%. 2,4,5It seems that precocious organ failure is determined by the extent to which necrosis affects the pancreatic parenchyma.In addition, organ failure diagnosed at admission increases the risk of MOF since 80% of the patients with acute severe pancreatitis characterised within 72 hours end up developing MOF despite the intensive care they receive.It should be noted that the progression of organ failure resulting from the disease itself during the first week of hospitalisation is associated with an increased mortality rate. 7n the present study, most deaths occurred during the first two weeks, the main cause being the systemic inflammatory response syndrome (69.5%).After this 14day period, the main cause of death was sepsis (87.5%).The main therapeutic problem regarding severe acute pancreatitis is the clinical intensive care for organ failure recovery. 4Despite the statement that deaths occurring in the early phase of ABP have been a rare event since the improvements in the intensive care unit, several studies have shown that many patients die during the first couple of weeks following hospitalisation. 6Patients with acute pancreatitis and MOF complication represent a clinical challenge in view of the lack of a specific therapy.It was shown that surgical intervention earlier than 2 weeks after the onset of symptoms is associated with a prohibitive mortality rate.Late interventions should be limited to those patients with necrosis infection, although the mortality rate remains elevated in the MOF cases regardless of the kind of surgical intervention. 6On the other hand, recent efforts have been made to improve the results using minimally invasive procedures or percutaneous drainage and antibiotics. 14he APACHE II score has been used during hospitalisation in order to predict the possibility of death based on the ROC curve.A sensitivity of 74.2% and specificity of 81.4% for APACHE score > 8.5 were obtained in the present study, which were higher than those found elsewhere, i.e., sensitivity ranging from 68% to 71% and specificity ranging from 48% to 67% for APACHE II score > 8 during hospitalisation. 15herefore, in spite of the fact that both medical regulation and SAMU have reduced the number of patients being admitted to emergency care units, and despite the technological and therapeutic resources, the mortality and severity rates for ABP cases have not been sufficiently reduced.Early identification of severe cases and immediate referral to an intensive care unit can open a new perspective in the evaluation of other therapeutic modalities for treating the acute inflammatory reaction, which probably will improve the prognosis for patients with ABP.