A simple scoring system for identifying patients with acute pancreatitis who do not require intensive care was 98% accurate and took about 30 minutes to complete, according to Dr. Paul Georg Lankisch and his colleagues.
The tool, known as the harmless acute pancreatitis score (HAPS), combines parameters that suggest a patient has a mild form of the disease: absence of rebound tenderness/guarding, normal levels of serum creatinine, and normal hematocrit.
“A physical examination of a patient with acute pancreatitis takes only minutes to find out whether he or she has rebound tenderness and/or guarding,” wrote Dr. Lankisch of the University of Göttingen (Germany) and his colleagues (Clin. Gastroenterol. Hepatol. 2009 [doi:10.1016/j.cgh.2009.02.020]). Hematocrit and serum creatinine are laboratory investigations available in every hospital at all times, and the findings are reported in about 30 minutes. Therefore, the HAPS yields a result in about a half-hour.
The authors pointed out that current tests used to determine the severity of pancreatitis “are insufficiently sensitive, too complicated, too expensive, and not available soon enough or not available at all outside specialized centers.”
Dr. Lankisch and his colleagues reported on findings from two prospective studies. In an effort to develop an easier way to identify patients with a first attack of acute pancreatitis who do not require intensive care, the researchers studied 394 patients with the condition who were admitted to the department of internal medicine at the Municipal Clinic in Luneburg, Germany, between 1987 and 2003.
A severe disease course was defined as presence of necrosis by contrast-enhanced CT (a Balthazar score of 5 or more points), while a nonsevere, “harmless” course was defined as having no necrosis (a Balthazar score of 0–4), no need for artificial ventilation or dialysis at any time during the hospital stay, and no fatal outcome.
Of the 394 patients, 143 had rebound tenderness and/or guarding and 251 did not. Baseline characteristics of all patients revealed that absence of rebound tenderness/guarding and normal serum creatinine levels were two strong predictors of a mild disease course. However, 23 of the 251 patients (9%) with no rebound tenderness/guarding and a normal serum creatinine level had a severe course.
The researchers observed that among these 251 patients, hematocrit levels exceeding 43% for men and 39.6% for women were strongly associated with having a severe course of the disease, so they added normal hematocrit levels as a third predictor to form the HAPS.
When applied to the 394 patients, the specificity of the HAPS was 97%, the sensitivity was 29%, the positive predictive value was 98%, and the negative predictive value was 22%.
Dr. Lankisch and his associates then sought to validate the HAPS in a multicenter study of 452 patients with a first attack of pancreatitis seen at one of three clinics between January 2004 and December 2006. These patients were similar to the initial set of patients in terms of pancreatic necrosis, and need for dialysis.
When applied to these patients, the specificity of the HAPS was 97%, the sensitivity was 28%, the positive predictive value was 98%, and the negative predictive value was 18%.
Combining results from the initial and validation sets of patients revealed that a severe course of disease was seen in only 4 of 204 patients (2%) whose pancreatitis was classified as harmless. “In two patients from the initial set and one from the validation set, the clinical condition deteriorated and for a short time, contrast-enhanced CT demonstrated small but definite areas of necrosis (Balthazar score 6 points),” the researchers reported.
“In the fourth patient, from the validation set, the acute pancreatitis healed without complications, but the patient died of methicillin-resistant Staphylococcus aureus pneumonia contracted during his hospital stay. Although the cause of death was unconnected with the pancreatitis, he was assigned to the 'not harmless' group because the fatal infection was acquired during his hospital stay for pancreatitis treatment.”
HAPS decides with great accuracy which patients' acute pancreatitis will run a mild course or who will have only interstitial pancreatitis, they concluded.
“Moreover, the score helps to decide which patients do not require intensive management and therapy and expensive imaging procedures, such as contrast-enhanced CT.”