About the CT Severity Index for Acute Pancreatitis
The CT Severity Index (CTSI) grades acute pancreatitis on contrast-enhanced CT by combining the Balthazar grade (0–4 points, for peripancreatic inflammation and fluid collections) with the extent of pancreatic necrosis (0–6 points), for a total of 0–10. A score of 0–3 is mild disease, 4–6 moderate, and 7–10 severe, with the derivation study reporting 92% morbidity and 17% mortality in the high-index group versus 2% morbidity and no deaths in the low-index group. It should be assessed on contrast-enhanced CT at 72 hours or later, since necrosis is not reliably visible earlier.
Formula
CTSI = Balthazar grade points (0–4) + necrosis points (0–6)- Balthazar grade points
- 0 (grade A) through 4 (grade E), from peripancreatic inflammation and collections.
- necrosis points
- 0, 2, 4 or 6, from the estimated percentage of non-enhancing gland.
- Necrosis points jump in steps of 2 (0, 2, 4, 6) rather than one point per category — a common source of manual scoring errors if the steps are assumed to be 0, 1, 2, 3.
- Necrosis is not reliably assessable before about 72 hours from symptom onset; scoring earlier can understate the eventual severity.
- The 0–10 total described here is the original Balthazar CT severity index. The modified CT severity index (2004) uses a different points structure and adds extrapancreatic complications — the two are not numerically interchangeable.
Interpreting the result
A low score (0–3) is reassuring and rarely needs repeat imaging unless the clinical course changes. A moderate score (4–6) predicts a longer admission and local complications, and should prompt closer clinical monitoring rather than routine repeat scanning. A high score (7–10) marks patients at genuinely elevated risk of infected necrosis, organ failure and death in the derivation cohort, and should trigger involvement of a pancreatic/intensive care service — but intervention decisions should be driven by clinical deterioration or confirmed infection, not by the imaging appearance of necrosis alone, since sterile necrosis is not itself an indication to drain.
| Score | Band | What it means | Action |
|---|---|---|---|
| 0–3 | Mild | Low morbidity and mortality in the derivation cohort | Supportive management; imaging rarely needs repeating unless the course changes |
| 4–6 | Moderate | Longer admission and higher rate of local complications | Anticipate complications; repeat imaging if fever, pain or organ dysfunction develops |
| 7–10 | Severe | 92% morbidity and 17% mortality in the high-index group of the derivation study, versus 2% morbidity and no deaths in the low-index group | Involve intensive care and a pancreatic service; do not drain sterile necrosis on imaging appearance alone |
Scroll the table sideways for every column.
What the CTSI needs (2 inputs)
- Balthazar grade
- A (normal pancreas, 0 points), B (focal or diffuse enlargement, 1 point), C (peripancreatic inflammation, 2 points), D (single fluid collection, 3 points), or E (two or more collections, or gas, 4 points).
- Pancreatic necrosis
- Estimated proportion of the gland failing to enhance: none (0 points), less than 30% (2 points), 30–50% (4 points), or more than 50% (6 points).
What it returns
- CTSI (0–10)
- Balthazar grade points plus necrosis points.
- Severity band
- Mild (0–3), moderate (4–6), or severe (7–10).
How it is calculated
Balthazar's original grading (A through E) captured peripancreatic inflammation and the presence of fluid collections, but on its own did not fully separate outcomes — the 1990 derivation work found that adding a specific assessment of pancreatic necrosis on contrast-enhanced CT sharpened that separation considerably, since non-enhancement of the gland is a direct sign of tissue death rather than an inference from inflammation alone. The two components are summed rather than combined by a formula, which keeps the score simple to apply at the scanner but means a patient can reach a high score either through extensive necrosis with a modest grade, or a severe grade with limited necrosis — two different pictures that land on the same number.
Facts & figures
| Group | Morbidity | Mortality |
|---|---|---|
| Necrosis present (n=22 of 88) | 82% | 23% |
| No necrosis | 6% | 0% |
| High CT severity index | 92% | 17% |
| Low CT severity index | 2% | 0% |
Scroll the table sideways for every column.
From the original 88-patient derivation cohort. The index was built precisely because necrosis and overall inflammatory grade together separated outcome far more sharply than either did alone.
Evidence
Derivation — Balthazar et al.
1990 · n = 8888 patients with acute pancreatitis assessed with bolus-injection dynamic contrast-enhanced CT at initial and follow-up examinations, evaluating the presence and degree of pancreatic necrosis (none, up to 30%, 30–50%, or over 50%) alongside peripancreatic inflammation and fluid collections.
Patients with necrosis had 23% mortality and 82% morbidity, versus 0% mortality and 6% morbidity without necrosis. Combining necrosis grading with the inflammatory grade into a CT severity index separated outcome further: patients with a high index had 92% morbidity and 17% mortality, versus 2% morbidity and no deaths in the low-index group.
Modified CT severity index — Mortele et al.
2004 · n = 6666 patients with contrast-enhanced CT within one week of symptom onset (from 266 consecutive acute pancreatitis admissions over one year), scored independently by three blinded radiologists using both the original CTSI and a modified index that simplifies necrosis/inflammation grading and adds extrapancreatic complications.
The modified index correlated more closely with length of stay, need for surgical or percutaneous intervention, and infection than the original CTSI, and was the only one of the two that correlated significantly with organ failure (p=0.0024 vs p=0.0513 for the original index). Interobserver agreement was similar between the two (kappa 0.71–0.85 modified vs 0.63–0.86 original).
How it compares
CTSI vs BISAP
Not alternatives on the same timeline — BISAP is calculated within the first 24 hours from clinical and laboratory data, while CTSI needs imaging that is most informative from 72 hours onward; many patients are appropriately scored with both, sequentially.
BISAP's advantage is availability at admission, before CT has necessarily been done or necrosis has had time to declare itself. CTSI's advantage is that it looks directly at the gland rather than inferring severity from systemic markers, which is precisely what a scan can show that blood tests cannot. A patient can have a reassuring early BISAP and a severe CTSI at 72 hours, or the reverse, and neither result overrides the other — they describe different stages of the same illness.
CTSI vs Modified CT severity index
The modified index has shown a closer correlation with patient outcome — particularly organ failure — than the original CTSI, in the study that introduced it, but the original remains widely used and the two point systems are not interchangeable.
The modified index simplifies the necrosis and inflammation grading and adds extrapancreatic complications (pleural effusion, ascites, vascular or gastrointestinal involvement) that the original CTSI does not capture at all. In a head-to-head reading of the same 66 patients, the modified index correlated significantly with organ failure where the original index did not reach significance, while interobserver agreement was similar for both.
Pearls & pitfalls
- Necrosis points step by twos (0, 2, 4, 6), not by one point per category — check the exact banding rather than assuming a linear 0–3 scale.
- Necrosis is frequently not visible before about 72 hours. An early CT can understate the eventual severity; repeat at 72 hours or sooner if the patient deteriorates.
- A high score does not by itself mean intervention is needed. Sterile necrosis on imaging is not drained on appearance alone — deterioration or confirmed infection drives that decision.
- The original 0–10 CTSI and the modified CT severity index are not the same points system; do not average or convert between them.
- CTSI is an imaging severity score, not a same-day admission triage tool — BISAP or the revised Atlanta criteria are what to use in the first 24 hours, before CT findings exist.
Critical actions
- Obtain contrast-enhanced CT at 72 hours or sooner if deteriorating, rather than immediately at admission, unless the diagnosis itself is in doubt.
- In a high-scoring patient, involve intensive care and a pancreatic surgery or interventional radiology service early.
- Do not treat visible necrosis alone as an indication for drainage; base intervention on clinical deterioration or confirmed infection.
- Repeat imaging if the clinical course changes — fever, worsening pain, or new organ dysfunction — rather than on a fixed schedule.
- Use a clinical score (BISAP, Ranson, revised Atlanta) for the admission-day severity question; CTSI answers a different, later question.
Why this score exists
Balthazar and colleagues were answering a specific question about prognosis rather than building a general severity scale: their 88-patient study set out to establish what contrast-enhanced CT adds, and it found that necrosis was the discriminator. Patients with necrosis had 23% mortality and an 82% complication rate; those without had no deaths and 6% morbidity. Only after establishing that did they combine peripancreatic inflammation, phlegmon and the degree of necrosis into an index, which separated outcomes starkly — 92% morbidity and 17% mortality at the high end against 2% morbidity and no deaths at the low end. The choice of collaborators tells the rest of the story: J. H. Ranson, whose clinical criteria were already the standard, was the senior author. The index was designed to add objective imaging information to clinical assessment rather than to displace it, which is why CTSI and the clinical scores are still used together rather than in competition.
About the creator
First author, 1990 derivation study
Established the CT grading of acute pancreatitis and the necrosis scoring that together form the index.
Senior author
The same Ranson whose clinical criteria predate this index — the CT severity index was developed explicitly to add imaging information to that clinical assessment.
Limitations
- Requires contrast-enhanced CT, with its associated radiation and contrast exposure, and is least accurate before roughly 72 hours from symptom onset.
- The necrosis-percentage estimate is a visual judgement by the reporting radiologist rather than a measured volume, which leaves room for interobserver variation.
- Does not capture extrapancreatic complications (pleural effusion, vascular involvement, bowel involvement) that the later modified index specifically added because they independently affect outcome.
- Derived in a modestly sized single-centre cohort (88 patients) by contemporary standards, though its core findings have been widely replicated in the decades since.
- Says nothing about organ failure directly measured at the bedside — it is an anatomical severity score, not a substitute for clinical assessment of the patient in front of you.
If you are the patient
The CT Severity Index is a score your radiology and gastroenterology team calculate from a CT scan of your pancreas, usually done a few days into a pancreatitis admission rather than on the first day, because some of what they are looking for — areas of the pancreas that have lost their blood supply — takes time to become visible. It combines how inflamed the area around the pancreas looks with how much of the gland shows this loss of blood supply, giving a score from 0 to 10. A low score is reassuring and rarely needs repeat scanning. A high score means a higher chance of complications and a longer stay in hospital, and your team will usually involve specialists in pancreatic disease, but a high score by itself does not automatically mean you need a procedure — that decision is based on whether you develop signs of infection or your condition worsens, not on the scan appearance alone.
Frequently asked questions
What is the CT Severity Index for pancreatitis?#
A 0–10 score combining the Balthazar grade (peripancreatic inflammation and fluid collections, 0–4 points) with the extent of pancreatic necrosis on contrast-enhanced CT (0–6 points). Higher scores predict more complications and higher mortality.
When should CT be done to calculate the CTSI?#
Ideally at 72 hours or later from symptom onset, since pancreatic necrosis is not reliably visible earlier. CT sooner may be needed if the patient is deteriorating, but a very early scan can understate the eventual severity.
What CTSI score is considered severe?#
7 to 10. In the derivation study, patients with a high CT severity index had 92% morbidity and 17% mortality, compared with 2% morbidity and no deaths in the low-index group.
Is the CT Severity Index the same as the modified CT severity index?#
No. They use different points systems. The modified index simplifies necrosis and inflammation scoring and adds extrapancreatic complications, and a comparison study found it correlated more closely with outcomes such as organ failure than the original CTSI.
Does a high CTSI mean I need drainage or surgery?#
Not by itself. Sterile necrosis visible on imaging is not an indication for intervention on its own — the decision to drain or operate is driven by clinical deterioration or confirmed infection, not by the CT appearance alone.
What is the difference between the Balthazar grade and the CTSI?#
The Balthazar grade (A to E) scores peripancreatic inflammation alone and converts to 0 to 4 points. The CTSI adds a necrosis component — 0 for none, 2 for up to 30%, 4 for 30 to 50%, and 6 for more than 50% — giving a total out of 10. So the Balthazar grade is one of the two components of the CTSI rather than a separate score, and quoting a Balthazar grade where a CTSI was requested omits the necrosis half of the assessment.
Does the CTSI require intravenous contrast?#
Yes for the necrosis component, which is the half that carries most of the prognostic weight. Necrosis is identified as non-enhancing pancreatic parenchyma, and that judgement cannot be made without contrast. A non-contrast study can support a Balthazar grade for peripancreatic inflammation but cannot produce a valid CTSI. Where contrast is contraindicated, MRI is the usual alternative.
How does the CTSI relate to the revised Atlanta severity classification?#
They describe different things on different timescales. The revised Atlanta classification grades severity by organ failure and its duration, and is usually settled by 48 hours; the CTSI grades what the scan shows, and necrosis is generally not radiologically evident until after 72 hours. So Atlanta severity is normally determined before a meaningful CTSI can be scored. The CTSI adds most where a patient has local complications without organ failure — moderately severe by Atlanta — because it quantifies the extent of necrosis that the single label does not convey.
References
Original / primary reference
Validation and refinement
Classification and guidelines
- Banks PA, Bollen TL, Dervenis C, et al. Classification of acute pancreatitis — 2012: revision of the Atlanta classification and definitions by international consensus. Gut. 2013;62(1):102-111.
- Working Group IAP/APA Acute Pancreatitis Guidelines. IAP/APA evidence-based guidelines for the management of acute pancreatitis. Pancreatology. 2013;13(4 Suppl 2):e1-e15.
- Tenner S, Vege SS, Sheth SG, et al. American College of Gastroenterology Guidelines: Management of Acute Pancreatitis. Am J Gastroenterol. 2024;119(3):419-437.