See “Serum Lipase as an Early Predictor of Severity in Pediatric Acute Pancreatitis” by Coffey et al on page 602.
Until recently, pancreatitis was considered an uncommon disease of childhood. That concept changed with the publication of a single-center study reporting an increased number of admissions for pediatric pancreatitis from 5 to 133 patients between 1993 and 1998 (1) and the subsequent publication of identical trends at other centers in the United States (2–4). The incidence of pancreatitis in children is now estimated to be around the incidence of pancreatitis in adults (∼1 in 10,000).
Although the pediatric acute pancreatitis is generally a self-limited condition, a subset of children develops severe pancreatic inflammation and necrosis, triggering systemic inflammatory response syndrome and multiorgan failure (mainly shock, renal failure, pulmonary insufficiency). The percentage of children who develop severe acute pancreatitis is variable in published series (6%–25%). This variation is probably the result of different etiologies, referral patterns, and expertise at various centers.
Because morbidity and the mortality increase sharply in patients with severe acute pancreatitis, scoring systems have been developed to predict the severity. This enables the physicians to identify patients at risk for developing severe complications, initiate close monitoring, fluid resuscitation, and increase the level of care as early as possible. Careful selection of patients and early intervention can improve the outcome from severe acute pancreatitis.
There are several scoring systems available to assess the severity of pancreatitis in adults, including Ranson, Glasgow, modified Glasgow, Bedside Index of Severity in Acute Pancreatitis, and APACHE II. Ranson criteria can be used on admission and within 48 hours to assess the prognosis as well. These scoring systems contain criteria (ie, age older than 55 years; fluid deficit >6 L) that are not easily applicable to children.
Pancreatitis scoring systems were first assessed in children with acute pancreatitis by DeBanto et al (5). They evaluated the performance of Ranson and modified Glasgow scores to predict the disease severity and compared with the criteria they have developed (DeBanto or pediatric acute pancreatitis score [PAPS]) (5). Their scoring system had 8 parameters, 4 to be scored at the time of admission and 4 by 48 hours. The admission criteria were as follows: age younger than 7 years, weight <23 kg, white blood cell (WBC) count >18,500 cells/mm3, and lactate dehydrogenase >2000 U/L. The 48-hour criteria were trough calcium <8.3 mg/dL, trough albumin <2.6 mg/dL, fluid sequestration >75 mL/kg every 48 hours, and an increase in blood urea nitrogen (BUN) >5 mg/dL. Like the Ranson criteria, 1 point was assigned for each criterion met. When the cutoff score for predicting a severe outcome was set at 3, the new system had a sensitivity of 70% versus 30% for Ranson and 35% for Glasgow scores. Negative predictive values (NPVs) were 91% versus 85% versus 85% and specificities were 79% versus 94% versus 94%. Positive predictive values (PPVs) of all of the tests were low: 45% versus 57% versus 61%. DeBanto et al concluded that their scoring system performed better than others because it had higher sensitivity and NPV. Later, 2 studies compared these 3 scores to predict the severity of pancreatitis in children and found that PAPS was not superior to others (6,7). The sensitivities of all of the scoring systems in predicting the disease severity were low (∼50%), NPVs were 80% to 90%, specificities 70% to 90%, and PPVs were ∼60% (6,7). One study found that admission WBC count >18,500 cells/mm3, trough calcium <8.3 mg/dL, and BUN increase >5 mg/dL were independent predictors of severe outcome in a logistic regression model (7). The major limitations of these studies were the low sensitivities and inability to calculate the scores at the time of admission (Ranson and PAPS).
In another study, Lautz et al (8) used contrast-enhanced computed tomographic images to predict the severity of pancreatitis in children. Known as the computed tomography severity index (CTSI) or Balthazar score (9), this scoring system is solely based on radiologic appearance, including pancreatic enlargement, peripancreatic inflammation, pancreatic fluid collection, and the extent of necrosis. Using a cutoff score ≥4, the sensitivity, specificity, PPV, and NPV of the CTSI were 81%, 76%, 62%, and 90%, respectively, which performed better than PAPS (53%, 72%, 41%, 80%), Ranson (71%, 87%, 67%, 89%), and modified Glasgow (71%, 87%, 67%, 89%) scores. Although CTSI score performed well, computed tomographies are not commonly ordered in children with acute pancreatitis. All of the acute pancreatitis scoring studies thus far have been done retrospectively.
In this issue of the Journal of Pediatric Gastroenterology and Nutrition, Coffey et al (10) propose a commonly used parameter, lipase, as a marker to segregate patients with severe acute pancreatitis within 24 hours of presentation. Lipase (>×7 upper limit of normal) as a scoring system for pediatric acute pancreatitis has its advantages because it is widely available, inexpensive, and easy to interpret. The ability to predict the severity of the disease within 24 hours of admission makes this score superior to other scoring systems, including Ranson and PAPS. The sensitivity and NPV of lipase were 85% and 89% respectively; therefore, lipase is a reasonable screening test for severe acute pancreatitis. The low PPV and specificity mean, however, that the test will identify many false-positives and has little clinical usefulness for the individual patient. The lipase score may be of potential use for future research and for the development of future disease activity indices and/or potential predictors for risk stratification.
In summary, we have made progress in understanding pediatric pancreatitis, its course, complications, treatment, and prognosis. Future challenges include developing solid criteria to define severe acute pancreatitis in children and form scoring systems that can predict the severity of disease at the time of presentation. This will allow us to identify children early, monitor them closely, and provide advanced care to improve outcome. In the future, we should consider conducting prospective studies in larger cohorts and developing pediatric scoring systems that will include clinical findings (eg, age-adjusted heart rate, blood pressure, respiratory rate, body mass index), laboratory parameters (eg, WBC count and differential, C-reactive protein, pCO2, BUN, creatinine), and imaging criteria (eg, fluid collections, organ enlargement, pancreatic necrosis).
1. Lopez MJ. The changing incidence of acute pancreatitis in children: a single-institution perspective. J Pediatr
2. Morinville VD, Barmada MM, Lowe ME. Increasing incidence of acute pancreatitis at an American pediatric tertiary care center: is greater awareness among physicians responsible? Pancreas
3. Park A, Latif SU, Shah AU, et al. Changing referral trends of acute pancreatitis in children: a 12-year single-center analysis. J Pediatr Gastroenterol Nutr
4. Werlin SL, Kugathasan S, Frautschy BC. Pancreatitis in children. J Pediatr Gastroenterol Nutr
5. DeBanto JR, Goday PS, Pedroso MR, et al. Acute pancreatitis in children. Am J Gastroenterol
6. Suzuki M, Fujii T, Takahiro K, et al. Scoring system for the severity of acute pancreatitis in children. Pancreas
7. Lautz TB, Chin AC, Radhakrishnan J. Acute pancreatitis in children: spectrum of disease and predictors of severity. J Pediatr Surg
8. Lautz TB, Turkel G, Radhakrishnan J, et al. Utility of the computed tomography severity index (Balthazar score) in children with acute pancreatitis. J Pediatr Surg
9. Balthazar EJ, Ranson JH, Naidich DP, et al. Acute pancreatitis: prognostic value of CT. Radiology
10. Coffey MJ, Nightingale S, Ooi CY. Serum lipase as an early predictor of severity in pediatric acute pancreatitis. J Pediatr Gastroenterol Nutr