Results

Cohort characteristics

A total of 58 adult patients with ACP were enrolled. The mean age was 43.9 years, with nearly half of the participants (n = 28; 48.28%) aged ≤ 40 years, indicating a substantial burden of disease among younger individuals. Males comprised 55.2% (n = 32) of the cohort, while females accounted for 44.8% (n = 26). Lifestyle analysis revealed alcohol consumption in 53.4% (n = 31) and smoking in 51.7% (n = 30), underscoring their prevalence as important background risk factors. The mean BMI was 25.03 ± 4.18 kg/m², placing the average participant in the overweight category, although values ranged from normal to obese. No significant correlation was observed between BMI and PCT levels. Patients were prospectively followed from emergency admission through serial 48-hour testing and assessment of final clinical outcomes at a tertiary care center (Table 1).

Parameter | Value
Age, years, mean ± SD | 43.9 ± 17.0
Gender
Male, n (%) | 32 (55.2%)
Female, n (%) | 26 (44.8%)
BMI, kg/m², mean ± SD | 25.03 ± 4.18
Alcohol consumption, n (%) | 31 (53.4%)
Pre-existing liver/kidney disease, n (%) | 34 (58.6%)

Clinical presentation

All patients presented with abdominal pain radiating to the back, a hallmark feature of ACP. Vomiting was reported in 51.7% (n = 30), while loss of appetite affected 53.4% (n = 31). Nausea was less frequent, occurring in 32.8% (n = 19). Imaging revealed necrosis in 20.7% (n = 12) of patients exceeding 30% and in 19.0% (n = 11) with < 30% necrosis. Ultrasound and magnetic resonance cholangiopancreatography (MRCP) identified calcifications in 31% (n = 18) and ductal strictures in 25.9% (n = 15), consistent with chronic disease changes (Table 2). Overall, 82.8% (n = 48) developed at least one local complication, ranging from acute peripancreatic fluid collections to walled-off necrosis, reflecting the high disease burden.

Parameter | Value
Abdominal pain, n (%) | 58 (100%)
Vomiting, n (%) | 30 (51.7%)
Serum amylase, U/L, mean ± SD | 1,678.22 ± 752.56
Serum lipase, U/L, mean ± SD | 2,312.19 ± 1,093.88
WBC count, /mm³, mean ± SD | 11,378.43 ± 3,801.29
Serum calcium, mg/dL, mean ± SD | 8.90 ± 0.93
Admission procalcitonin, ng/mL, mean ± SD | 2.64 ± 1.47
48-hour procalcitonin, ng/mL, mean ± SD | 2.48 ± 1.51
Pancreatic necrosis > 30% (CT), n (%) | 12 (20.7%)
Pancreatic calcifications (USG), n (%) | 18 (31.0%)

Baseline laboratory findings

Admission biochemistry demonstrated marked derangements. Mean serum amylase was 1678.22 ± 752.56 U/L, with only 24.1% (n = 14) of patients having values within normal limits. Lipase was similarly elevated (mean 2312.19 ± 1093.88 U/L). The mean WBC count was 11,378.43/mm³, with 72.4% (n = 42) showing leukocytosis. Admission PCT levels varied widely (mean 2.64 ± 1.47 ng/mL; range 0.14-4.99 ng/mL). Additional abnormalities included borderline hypocalcemia (mean 8.90 ± 0.93 mg/dL) and raised creatinine levels (mean 1.54 ± 0.53 mg/dL), indicating early metabolic and renal dysfunction in a subset of patients (Table 2).

Serial PCT trends

Dynamic changes in PCT provided critical prognostic insights. At admission, the mean PCT level was 2.64 ± 1.47 ng/mL. After 48 hours, the mean level declined slightly to 2.48 ± 1.51 ng/mL, although the overall values concealed divergent trajectories. Patients progressing to severe illness or death exhibited persistently elevated or rising PCT levels, whereas those who improved clinically showed a steady decline. This pattern demonstrated that serial PCT measurements, rather than a single baseline value, were more predictive of disease progression and complications (Figure 2).

Figure 2: Serial procalcitonin trends

Clinical outcomes

Morbidity and mortality were considerable. Seven patients (12.1%; n = 7) died during hospitalization (Figure 3).

Figure 3: Clinical outcomes and causes of mortality in acute-on-chronic pancreatitisARDS: acute respiratory distress syndrome; MODS: multiple organ dysfunction syndrome

ICU admission was required in 60.3% (n = 35) of patients due to organ failure or systemic complications. Local complications were frequent (82.8%; n = 48), including peripancreatic fluid collections (22.4%; n = 13), necrosis (20.7%; n = 12), and walled-off necrosis (15.5%; n = 9). Length of stay varied widely, influenced by disease severity and comorbidities. Survivors accounted for 87.9% (n = 51) of patients at discharge, while patients with multiple organ failure (Modified Marshall Score > 3) had significantly worse outcomes (p = 0.0003), highlighting the complexity of ACP management (Table 3).

Outcome | Normal PCT (n = 9), n (%) | Elevated PCT (n = 49), n (%) | P-value
Acute organ failure | 2 (22.2%) | 29 (59.2%) | 0.021
Local complications | 2 (22.2%) | 32 (65.3%) | 0.041
ICU admission | 5 (55.6%) | 30 (61.2%) | 1.000
Mortality | 0 (0.0%) | 7 (14.29%) | 0.79

Association between PCT and outcomes

Higher PCT levels were strongly associated with adverse outcomes. Patients admitted to the ICU had significantly elevated mean PCT levels (2.95 ± 1.45 ng/mL) compared with those managed in the ward (2.20 ± 1.50 ng/mL; p = 0.034). Although admission PCT did not directly predict mortality, non-survivors consistently had higher values. Strong correlations were observed between PCT levels and length of hospital stay (r = 0.969 at admission; r = 0.979 at 48 hours; p = 0.0001). Elevated PCT was also associated with acute organ failure (p = 0.021) and local complications (p = 0.041), confirming its role as a marker of systemic inflammatory escalation (Figure 4).

Figure 4: Association between PCT levels and clinical severityPCT: procalcitonin; ICU: intensive care unit

ROC analysis

ROC analysis highlighted the superiority of 48-hour PCT over admission values. Admission PCT yielded poor discriminatory ability (AUC = 0.555; sensitivity = 46.9%; specificity = 76.9% at a cutoff of 3.46 ng/mL). In contrast, 48-hour PCT achieved excellent accuracy (AUC = 0.992). At a threshold of 4.04 ng/mL, sensitivity reached 100% with a negative predictive value (NPV) of 100%, specificity was 92.3%, and the good predictive value was 94.1%. These findings establish 48-hour PCT as a reliable predictor of mortality and severe morbidity (Table 4) (Figure 5).

Marker | AUC | Best threshold | Sensitivity | Specificity | NPV
Admission PCT | 0.555 | 3.46 ng/mL | 46.9% | 76.9% | 54.1%
48-Hour PCT | 0.992 | 4.04 ng/mL | 100% | 92.3% | 100%

Figure 5: ROC curveROC: receiver operating characteristic; PCT: procalcitonin; AUC: area under the curve

Mortality

Among the seven patients who died during hospitalization, acute respiratory distress syndrome (ARDS) was the most common cause of mortality, accounting for three deaths (42.9%; n = 3). Multiple organ dysfunction syndrome (MODS) was responsible for two deaths (28.6%; n = 2). One patient died due to infected pancreatic necrosis complicated by sepsis, while another succumbed to septic shock. These findings indicate that mortality in ACP was primarily driven by severe systemic inflammatory and infectious complications, leading to organ dysfunction and failure.

Cause of death | N | % | Average PCT value (= 48 hrs)
ARDS | 3 | 42.9 | 7.28
MODS | 2 | 28.6 | 6.77
Infected pancreatic necrosis with sepsis | 1 | 14.3 | 6.24
Septic shock | 1 | 14.3 | 6.19
Total | 7 | 100.0 |