Comparative Study of ED-SAS and Harmless Acute Pancreatitis Score (HAPS) for Predicting ICU Admission in Patients with Acute Pancreatitis in the Emergency Department
Open Access Original Research Paper Comparative Study of ED-SAS and Harmless Acute Pancreatitis Score (HAPS) for Predicting ICU Admission in Patients with Acute Pancreatitis in the Emergency Department Authors: Dr. Vipul Choudhary¹, Dr. Jayant Makkar², Dr. Shuchita Vaya³, Dr. Naveen Goyal⁴, Dr. Revant Aggarwal⁵, Dr. Avniendra Bharadawaja⁶ ¹Senior Resident, Department of Emergency Medicine, Prince Medical College and Hospital, Sikar, Rajasthan, India. ²Senior Resident, Department of Emergency Medicine, Prince Medical College and Hospital, Sikar, Rajasthan, India. ³Professor and Head, Department of Emergency Medicine, Mahatma Gandhi Medical College, Jaipur, Rajasthan, India. ⁴Professor, Department of Emergency Medicine, Mahatma Gandhi Medical College, Jaipur, Rajasthan, India. ⁵Assistant Professor, Department of Emergency Medicine, Mahatma Gandhi Medical College, Jaipur, Rajasthan, India. ⁶Senior Resident, Department of Emergency Medicine, Mahatma Gandhi Medical College, Jaipur, Rajasthan, India. Corresponding Author: Dr. Jayant Makkar*, Senior Resident, Department of Emergency Medicine, Prince Medical College and Hospital, Sikar, Rajasthan, India. DOI: https://doi.org/10.5281/zenodo.24270968 Received: 03 October 2026, Revised: 18 October 2026, Accepted: 04 November 2026, Online First: 18 November 2026 ABSTRACT Background: Early identification of acute pancreatitis (AP) patients who will need intensive care is difficult in the emergency department (ED), because classic scores take 24–48 hours. The Harmless Acute Pancreatitis Score (HAPS) is a rule-out tool, and the ED-SAS score (SpO₂, age, SIRS) is a bedside tool built for mortality. We compared both, and a combination of the two, for predicting ICU admission. Methods: This prospective observational study enrolled 47 adults with AP at a tertiary care ED in Jaipur, India (April 2024 to September 2025). ED-SAS (age >60 years, SpO₂ <96%, ≥2 SIRS criteria; high risk ≥2) and HAPS (rebound tenderness or guarding, creatinine ≥2 mg/dL, haematocrit above sex-specific cut-offs) were calculated at presentation. A combined score classified patients as high risk (both positive), low risk (both negative) or moderate risk (discordant). The primary outcome was ICU admission; the secondary outcome was mechanical ventilation. Results: Mean age was 42.4 ± 14.9 years and 91.5% were male. Sixteen patients (34.0%) were admitted to ICU and four (8.5%) were ventilated. For ICU admission, ED-SAS had sensitivity 87.5%, specificity 83.9%, PPV 73.7%, NPV 92.9% (AUC 0.857). HAPS had sensitivity 93.8%, specificity 54.8%, PPV 51.7%, NPV 94.4% (AUC 0.743). The combined high-risk category had sensitivity 81.3%, specificity 87.1%, PPV 76.5%, NPV 90.0%. ICU admission was 76.5% in the high-risk group, 21.4% in the moderate-risk group and 0% in the low-risk group. All four ventilated patients were in the combined high-risk group. Conclusion: ED-SAS discriminated ICU need well, HAPS acted mainly as a sensitive rule-out test, and combining them gave the highest specificity and PPV with a graded three-tier triage. These findings come from a small single-centre cohort and need multicentre validation. Keywords: acute pancreatitis; emergency department; ED-SAS; HAPS; ICU admission; risk stratification INTRODUCTION Acute pancreatitis (AP) is a common cause of emergency admission, and its course ranges from self-limiting interstitial inflammation to necrotizing disease with persistent organ failure.[1,2] Incidence is reported at 13–45 per 100,000 population per year in India, with gallstones and alcohol the dominant causes, and about 15–20% of patients develop moderately severe or severe disease.[3,4] Mortality rises to 15–30% in severe disease, and most of it is driven by persistent organ failure.[1,5] The first 24–48 hours largely determine the course. During this window the emergency department (ED) must decide who needs resuscitation, monitoring and ICU referral, and who can safely be managed on a ward.[2,5] The Revised Atlanta Classification defines severity by organ failure and local complications, but it is applied retrospectively and does not support triage at first contact.[1] Ranson's criteria, the Glasgow score and APACHE II need 24–48 hours of data or many variables, and BISAP needs laboratory results and imaging that may not be available at presentation.[6–8] Two simple bedside tools have been proposed for the ED. The Harmless Acute Pancreatitis Score (HAPS) combines absence of rebound tenderness or guarding, serum creatinine below 2 mg/dL and a normal haematocrit (below 43% in men, 39.6% in women). It was designed to identify non-severe disease and has a high negative predictive value.[9,10] The ED-SAS score uses age above 60 years, SpO₂ below 96% and two or more SIRS criteria, all available within minutes of arrival, and was derived to predict mortality.[11,12] HAPS captures early organ dysfunction and haemoconcentration, whereas ED-SAS captures systemic inflammation and oxygenation, so the two may complement each other. Direct comparisons of these two scores are scarce, particularly in Indian EDs, and their combined use has hardly been studied. We therefore compared ED-SAS, HAPS and a combined score for predicting ICU admission in adults with AP presenting to the ED, with mechanical ventilation as a secondary outcome. METHODS Design and setting This was a hospital-based prospective observational comparative study in the Emergency Department of Mahatma Gandhi Medical College and Hospital, Jaipur, a tertiary care teaching hospital with 24-hour emergency services. Enrolment ran for 18 months, from April 2024 to September 2025, after approval from the Institutional Ethics Committee. Participants We included patients aged 18 years or older who were clinically diagnosed with AP on clinical features and/or imaging and who gave written informed consent. We excluded chronic pancreatitis, traumatic pancreatitis, post-operative or post-ERCP pancreatitis, pancreatic malignancy, and pregnant or immunocompromised patients. Forty-seven patients were enrolled consecutively. Data collection At presentation we recorded heart rate, respiratory rate, blood pressure, temperature and SpO₂, performed a clinical abdominal examination, and obtained serum creatinine, haematocrit and total leukocyte count. All scores were calculated at ED presentation, before the ward or ICU decision. Treatment followed standard hospital protocol and the study did not influence it. Scoring systems SIRS was defined as two or more of: temperature below 36 °C or above 38 °C; pulse above 90/min; respiratory rate above 20/min or PaCO₂ below 32 mmHg; WBC above 12,000/mm³, below 4,000/mm³ or more than 10% immature bands. ED-SAS gave one point each for age above 60 years, SpO₂ below 96% and two or more SIRS criteria (range 0–3). A score of 2 or more was high risk and 0–1 was low risk. HAPS gave one point for each of: rebound tenderness or guarding present; serum creatinine 2 mg/dL or higher; haematocrit at or above 43% (men) or 39.6% (women). A score of 0 was HAPS-negative (predicts non-severe disease) and 1 or more was HAPS-positive. Combined score: high risk = ED-SAS high risk and HAPS-positive; low risk = ED-SAS low risk and HAPS-negative; moderate risk = discordant results. For two-by-two analysis, the high-risk group was test-positive and the moderate and low groups were test-negative. Outcomes The primary outcome was ICU admission, used as a surrogate marker of severe disease. The secondary outcome was mechanical ventilation. Statistical analysis Data were analysed in Microsoft Excel and SPSS. Categorical variables are shown as counts and percentages and continuous variables as mean ± SD. Sensitivity, specificity, PPV, NPV and accuracy were calculated from two-by-two tables with ICU admission as the reference standard. Associations were tested with the chi-square test, and p < 0.05 was taken as significant. Areas under the ROC curve (AUC) were calculated for each score. Ethics The Institutional Ethics Committee of Mahatma Gandhi Medical College and Hospital, Jaipur, approved the study (ref. MGMC&H/IEC/JPR/2024/3027, 26 April 2024). Written informed consent was obtained from all participants, confidentiality was maintained, and no intervention beyond standard care was performed. RESULTS Study population Forty-seven patients were analysed. Mean age was 42.4 ± 14.9 years (range 20–80), and 43 (91.5%) were male, a male-to-female ratio of about 10.8:1. Gallstones were the commonest cause (53.2%), followed by alcohol (36.2%). Sixteen patients (34.0%) were admitted to ICU and four (8.5%) needed mechanical ventilation. Table 1. Baseline characteristics (n = 47) Characteristic Value Age, years, mean ± SD 42.4 ± 14.9 Age >60 years, n (%) 8 (17.0) Male sex, n (%) 43 (91.5) Gallstones / alcohol / other, n (%) 25 (53.2) / 17 (36.2) / 5 (10.6) Pulse, /min, mean ± SD 102.5 ± 16.2 Respiratory rate, /min, mean ± SD 18.0 ± 2.1 Systolic BP, mmHg, mean ± SD 117.0 ± 21.4 Temperature, °C, mean ± SD 37.77 ± 0.77 SpO₂, %, mean ± SD 95.6 ± 4.3 Total leukocyte count, /mm³, mean ± SD 13,043 ± 4,324 Serum creatinine, mg/dL, mean ± SD 1.69 ± 1.62 Haematocrit, %, mean ± SD 41.3 ± 6.5 ICU admission, n (%) 16 (34.0) Mechanical ventilation, n (%) 4 (8.5) Score distributions at presentation On ED-SAS, 19 patients (40.4%) were high risk (score ≥2): 15 patients (31.9%) had SpO₂ below 96% and 28 (59.6%) had two or more SIRS criteria. ED-SAS scores of 0, 1, 2 and 3 were seen in 18, 10, 16 and 3 patients. On HAPS, 29 patients (61.7%) were positive; 16 (34.0%) had rebound tenderness or guarding, 9 (19.1%) had creatinine of 2 mg/dL or more and 25 (53.2%) had a raised haematocrit. HAPS scores of 0, 1, 2 and 3 were seen in 18, 12
Authors
- Vipul Choudhary
- Jayant Makkar
- Naveen Goyal⁴
- Shuchita Vaya³
- Avniendra Bharadawaja⁶
- Revant Aggarwal⁵
Publication Details
- Journal
- International Journal of Medical Science in Clinical Research and Review
- Published
- 2026-11-18
- DOI
- https://doi.org/10.5281/zenodo.24270967
- Primary Topic
- Pancreatitis Pathology and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00