Modified Marshall Score: An Underutilised Prognostication Tool for Acute Pancreatitis

Introductions Acute pancreatitis (AP) remains an increasing cause of mortality in the United Kingdom (UK) with a multitude of aetiologies. Nevertheless, it is known that organ failure due to systemic injury occurs as a complication in the early stage of 20% of AP cases. Early detection and proper risk stratification are paramount for clinicians to make informed decisions. The modified Marshall (MM) score is the default risk stratification tool to assess organ dysfunction in acute pancreatitis patients in local practice. Aims/Methods This review retrospectively evaluated the utilisation of the Modified Marshall score amongst acute pancreatitis patients admitted under the General Surgery department of a local district hospital over a period of two years. Our primary goal was to determine the proportion of patients with a completed Modified Marshall score during admission clerking. For our secondary goals, we then risk-stratified each AP case and compared it with the actual patient outcome. Electronic records, discharge letters, and surgical handover sheets were thoroughly assessed to facilitate the data collection process. A questionnaire, which further explored the reasons for underutilisation, was sent to all clerking clinicians and practitioners. Results A total of 127 acute pancreatitis patients were admitted with a mean age of 57.1±16.6 years, a 1.4:1 male-to-female ratio, and a median serum amylase of 377±18. Out of 127 cases, only 24 (18.9%) patients were risk-stratified using the Modified Marshall score. We received 19/21 survey responses. Common reasons for underutilisation were unfamiliarity with scoring components, utilisation complexity involving unit conversion, work demand, and preference over other scoring systems. To address the issues, we simplified the MM score formula and designed new unit conversion guidance for the fractional inspired oxygen (FiO₂) component. Following the implementation of our new measure, another review measured an improvement in MM score utilisation over a month in 7/10 patients. From our retrospective risk stratification, the proportion of patients being risk stratified as severe pancreatitis was 6.30%, 9.45% and 0.79%, respectively, via the Modified Marshall score, Glasgow-Imrie score, and CT severity index. Amongst those scoring low Modified Marshall score <2 (119 patients), 95.8% (114) patients did not progress to requiring intensive therapy unit (ITU) admission or emergency exploratory surgery or drainage. Conclusion The modified Marshall score is currently underutilised in our local clinical practice. Although scoring systems are deemed less superior than the evaluation of experienced clinicians, the MM score remains a reliable scoring system with high negative predictive value (NPV) to help surgical trainees to objectively risk stratify AP cases and facilitate more appropriate ITU referrals.