A Clinical and Experimental Study of Intraperitoneal Antiprotease Therapy in Acute Pancreatitis

Wilson, Colin (1989) A Clinical and Experimental Study of Intraperitoneal Antiprotease Therapy in Acute Pancreatitis. MD thesis, University of Glasgow.

Abstract

Peritoneal exudate appears to be toxic in experimental pancreatitis, possibly as a result of overwhelming of its antiprotease defences by proteolytic enzymes released from the pancreas. Removal of this exudate by peritoneal lavage has been a uniformly successful therapy. In contrast, the role of the protease-antiprotease balance in peritoneal exudate in human acute pancreatitis, and its relationship with the early "shock-like" illness which may complicate severe attacks, has not been clearly defined and the efficacy of peritoneal lavage is unproven. This thesis has addressed 3 main issues. The nature and extent of the problem presented by acute pancreatitis has been investigated, as have means for monitoring and predicting the severity of the illness. The major part of the thesis has examined aspects of the protease-antiprotease balance in the peritoneal exudate that complicates severe acute pancreatitis, and the efficacy of treatment by the administration of intraperitoneal antiproteases in experimental pancreatitis and in man. Incidence and mortality The incidence and mortality trends of acute pancreatitis in Scotland have been reviewed between 1961 and 1985 using data from the Scottish Hospital In-patient Statistics. Over this 25 year period the number of patients discharged with a diagnosis of acute pancreatitis increased 11-fold in males and 4-fold in females and may be largely explained by an increased frequency of diagnosis (diagnostic rate), due to greater clinical awareness and more frequent diagnostic testing. The absolute mortality rate has increased slightly although, because of the marked increase in incidence, the case mortality rate has fallen from 17.8% to 5.6%. The incidence and mortality trends were also examined in Glasgow Royal Infirmary between 1974 and 1984, over which period the diagnostic rate was thought to have been unchanged. Despite a 23% fall in the catchment population the annual number of admissions remained relatively constant, suggesting a true increase in local incidence. Amongst the fatal attacks, 42% had the diagnosis first made at post mortem, the fall in overall mortality from 14.9% to 10.8% reflecting a reduction in the numbers first diagnosed at post mortem. Mortality amongst patients diagnosed in life was unchanged at 9% throughout, thus when the effect of an increased diagnostic rate is excluded, improvements in therapy do not appear to have influenced the mortality rate. Prediction of outcome Investigation of invasive or expensive therapies requires that patients with severe pancreatitis and at risk of a complicated attack, are identified accurately, early in their illness. There is also a need to develop means to objectively monitor the course of the illness and any response to therapy, which might aid their management. Simple clinical assessment was shown to have the highest overall accuracy in predicting outcome but, even 48 hours post-admission, detected under 50% of those developing complications. Multiple factor scoring systems provided a useful prediction of outcome, particularly for attacks associated with alcohol abuse, but provided a "once only" prediction and failed to allow for sequential monitoring of the course of an attack. Sequential monitoring of 5 different complement factors was of no value in discriminating complicated from uncomplicated attacks. Serum antiproteases demonstrated a falling alpha2macroglobulin concentration and a rising alpha1antiprotease concentration in complicated attacks, both providing useful discrimination. C-reactive protein provided better discrimination but peak levels (>210mg/1 on the 2nd, 3rd or 4th day), while providing equivalent accuracy to the multiple factor scoring systems incurred a similar delay. C-reactive protein concentrations fell with clinical improvement, persistently high levels (>120mg/1 on day 7) indicating an increased risk of complication. The APACHE II scoring system provided a prediction of outcome within a few hours of admission. The peak score recorded in the first 3 days provided greater accuracy, equivalent to the multiple factor scoring systems, but incurred a similar delay. The APACHE II was objective, reproducible and appeared to reflect improvement or deterioration in the patient's clinical condition and may permit sequential monitoring of the course of the illness. None of the scoring systems or factors examined was able to predict complicated attacks earlier and more accurately than multiple factor scoring systems. C-reactive protein and APACHE II both appear to be of value, C-reactive protein for its simplicity and both for objective monitoring during the course of an attack. Intraperitoneal antiprotease therapy Retrograde intraductal injection of sodium taurocholate produced a severe pancreatitis in rats, death usually occurring within 24 hours. Only a small proportion of patients die of such a fulminant, rapidly fatal illness but it is this early "shock-like" illness which therapy aims to ameliorate and, therefore, this model appeared an appropriate one to study. (Abstract shortened by ProQuest.).

• Item Type: Thesis (MD)
• Qualification Level: Doctoral
• Keywords: Medicine, Pharmacology
• Date of Award: 1989
• Depositing User: Enlighten Team
• Unique ID: glathesis:1989-77906
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 28 Feb 2020 12:09
• Last Modified: 28 Feb 2020 12:09
• URI: https://theses.gla.ac.uk/id/eprint/77906