Clinicopathological characteristics of pancreatitis in Far North Queensland

Pancreatitis is a common cause for unscheduled admissions in General or Gastrointestinal Surgery units throughout the world. This is particularly perceived to be the case in northern Australia, where alcohol and certain socio-demographic factors may be prominent in influencing the incidence and course of the disease. Over the years, much has been written about various aspects of pathogenesis, prognostication and management. However in the average clinical setting, many useful questions go unasked and unanswered. Because patients rarely require formal surgical intervention or even high-dependency management, there is rarely any active enquiry into the issues surrounding acute pancreatitis admissions. The chronic aspects of what is superficially considered to be an isolated acute process are also often overlooked.\ud \ud This body of work evolved out of the desire to foster a better standard of care for such patients and ultimately to reduce the frequency of acute hospital admissions, particularly recurrent admissions for the same patient.\ud \ud In order to effect any improvement in the management of a disease, it is first necessary to quantify and characterize it in relation to the target population. To this end, a retrospective audit was undertaken of acute pancreatitis admissions to the three major referral hospitals of Northern Queensland over the calendar year of 1997. Any inferences from this study were limited by the quality of the data recorded by a variety of clinicians. However, from what was available, it was evident that many patients admitted to hospital with a diagnosis of acute pancreatitis appeared to exhibit the characteristics of a chronic disease process. Indeed almost half of those who fulfilled the diagnostic criteria for acute pancreatitis appeared to be harbouring underlying chronic pancreatitis, based on clinical specifications.\ud \ud To better characterize the clinicopathological features of hospitalizations for pancreatitis in Northern Queensland, it was therefore necessary to acquire case data in a robust and prospective manner. A purposive and iterative literature review was undertaken to identify gaps in existing knowledge and areas that would merit further enquiry. This also assisted in informing the explanatory variables that would constitute the prospective data collection. Thus conceived was the accrual of a cohort of patients presenting acutely to Cairns Base Hospital in Far North Queensland.\ud \ud Regarding outcome measures, a more objective means of diagnosing chronic pancreatitis was deemed to be desirable. The newly available faecal elastase-1 assay ([FE-1]) was proposed as a means of augmenting diagnostic certainty for chronic pancreatitis, as defined by exocrine insufficiency. The accuracy of [FE-1] in the study population was validated by a pilot study of acute hospitalizations. At the inception of the study, this assay was not available within the public health sector and was therefore performed for all cases by the researcher. The pilot study found [FE-1] to have highly acceptable positive predictive value for diagnosing underlying chronic pancreatitis in acute admissions that exhibited no evidence of developing severe acute pancreatitis.\ud \ud For the eventual cohort study, the feasibility of obtaining stool specimens for all acute hopsitalizations, many of whom were admitted to hospital for less than three days, meant that [FE-1] could not be relied upon to diagnose all cases of chronic pancreatitis, if maximal patient recruitment were desired. A composite case definition that incorporated [FE-1] and a number of validated clinical parameters were therefore utilized to define this sub-group within the study population.\ud \ud For hospitalizations fulfilling diagnostic criteria for acute pancreatitis, a variety of other explanatory and outcome variables were collected. From March 2004 to July 2007, 153 cases were recruited for prospective data collection. The mean age of those admitted was 44.7 years. Males accounted for 61.4% of the cohort (n=94) and 41.2% (n=63) identified as Indigenous. In 56.9% (n=83), aetiology was deemed to be alcohol-related and in 24.7% (n=36) it was biliary or gallstone-related. Eight-three cases (54.3%) fulfilled the composite diagnostic criteria for chronic pancreatitis, and 11.2% (n=17) in the course of their admission developed severe acute pancreatitis according to the Atlanta definition.\ud \ud The prospectively acquired data of the patient cohort recruited between March 2004 and July 2007 were further analysed to a number of ends:\ud \ud Firstly, the incidence or admission rate of acute pancreatitis in the Far North Queensland population was estimated using a capture-recapture method, in order to account for cases missed by study recruitment or hospital separation records. This revealed that the likely admission rate for acute pancreatitis was in the order of 16.1 admissions per month, which would approximate a crude annual incidence (including recurrent admissions) of 84 per 100 000. It was therefore evident that the true incidence of the disease may be considerably underestimated by the traditional means of enumeration.\ud \ud Secondly, cross-sectional associations with chronic pancreatitis were sought using a variety of available explanatory variables. Nutrient intakes were of particular interest, but none of the decomposed micro- or macronutrients showed a significant association. When these were recast as patterns of exogenous intakes using Principal Components Analysis, it emerged that patients with underlying chronic pancreatitis, in the 24 hours prior to the onset of an acute exacerbation, were characterized by an avoidance of foodbased nutrients in favour of non-nutritive substances, such as coffee and tobacco. This appeared to be independent of alcohol intake.\ud \ud Thirdly, novel clinical determinants of severe acute pancreatitis were explored, exploiting the forward directionality of explanatory and outcome variables. This analysis confirmed the causal association with central adiposity inferred by a number of other studies. It also suggested an unprecedented negative association with smoking. This would merit corroboration by other observational studies, as well basic scientific research to elucidate possible mechanisms such as activation of a nicotinic anti-inflammatory pathway. Moreover, the identification of modifiable determinants of severe acute pancreatitis may contribute to future preventive and therapeutic strategies.\ud \ud In conclusion, the picture that has emerged from the Far North Queensland cohort is that hospitalizations for pancreatitis, while by definition classified as acute, may in a broader sense exhibit the characteristics of a chronic disease process. Such a reconceptualization of the disease process may serve to inform more effective models of care that are tailored to both the clinicopathological characteristics of the disease and the specific population that it affects