Ulcer classification refers to a family of scoring and staging systems that doctors use to grade the severity, depth, cause, and bleeding risk of ulcers across the body. There is no single universal scheme. Instead, different specialties have developed their own frameworks: gastroenterologists grade bleeding peptic ulcers with the Forrest classification, wound-care teams stage pressure injuries from I to IV, vascular surgeons apply the CEAP system to venous leg ulcers, and diabetic foot specialists choose among at least five competing tools. Understanding which system applies, and what its grades actually mean, matters because it drives treatment decisions ranging from whether an endoscopy needs intervention to whether a limb can be saved.
Bleeding Peptic Ulcers and the Forrest Classification
When a gastroenterologist looks at a stomach or duodenal ulcer through an endoscope, the first question is whether it is actively bleeding and how likely it is to bleed again. The Forrest classification, introduced in 1974, remains the standard tool for answering that question. It divides findings into three main groups based on what the endoscopist sees on the ulcer surface.
Forrest I ulcers are actively bleeding. Within that group, Ia means arterial spurting and Ib means oozing. Forrest II ulcers show signs of recent bleeding but are not actively hemorrhaging at the moment: IIa has a visible vessel on the ulcer base, IIb has an adherent blood clot, and IIc has a dark spot of flat pigment. Forrest III is a clean-based ulcer with no stigmata of bleeding at all. The clinical value of this grading is straightforward: higher Forrest classes carry higher rebleeding risk and generally require endoscopic treatment, while lower ones can often be managed with medication alone.
A large international trial quantified rebleeding rates for each category in patients who received placebo after endoscopic treatment. Spurting arterial bleeding (Forrest Ia) carried a rebleeding rate of about 23%, adherent clots (IIb) around 18%, visible vessels (IIa) roughly 11%, and oozing (Ib) only about 5%.1PubMed Central. Reassessment of Rebleeding Risk of Forrest IB (Oozing) Peptic Ulcer Bleeding in a Large International Randomized Trial That last finding was somewhat surprising: oozing ulcers had long been grouped with other active bleeders, but their rebleeding risk turned out to be far lower, closer to the low-risk categories than to arterial spurters. High-dose acid-suppressing medication after endoscopy reduced rebleeding for the higher-risk categories but made little difference for oozing ulcers, whose rates were already low.
Image-analysis research has added a visual dimension to these grades. Active Forrest I lesions showed the largest area of bleeding on endoscopic images, while clean-based Forrest III ulcers had none. Interestingly, adherent clots (IIb) covered a larger proportion of the image than visible vessels (IIa), which may help explain why clot-covered ulcers rebleed more often despite appearing more “sealed.”2PubMed Central. Forrest Classification for Bleeding Peptic Ulcer: A New Look at the Old Endoscopic Classification
Pressure Injury Staging
Pressure injuries, still commonly called bedsores or pressure ulcers, are staged by depth of tissue damage. The National Pressure Ulcer Advisory Panel (NPUAP) system is the most widely used framework and grades injuries from Stage 1 through Stage 4, with two additional categories for special situations.
Stage 1 is intact skin with non-blanchable redness over a bony prominence. You press on it and the redness does not fade. The skin may feel warmer, cooler, firmer, or more painful than the surrounding area. This stage can be difficult to detect in people with darker skin tones, where color changes may be subtle.3Journal of the American College of Clinical Wound Specialists. The NPUAP Meeting – This was No Consensus Conference – Section: Stage I Pressure Ulcer/Stage 1 Pressure Injury Stage 2 involves partial-thickness skin loss: you see a shallow, open wound with a pink-red bed, or sometimes a blister that is intact or ruptured. There is no dead tissue (slough) visible.
Stage 3 is full-thickness skin loss where you can see subcutaneous fat, but bone, tendon, and muscle remain covered. Slough may be present, and the wound may tunnel beneath the skin surface. Stage 4 is the most severe: full-thickness tissue loss with exposed bone, tendon, or muscle, often with undermining and tunneling.4Journal of the American College of Clinical Wound Specialists. The NPUAP Meeting – This was No Consensus Conference – Section: Stage IV Pressure Ulcer The depth of both Stage 3 and Stage 4 injuries varies by body location: a Stage 3 ulcer on the bridge of the nose, which has little subcutaneous tissue, looks very different from one on the buttock.
Two additional categories handle situations where staging by depth is not possible. “Unstageable” injuries are those where slough or dead tissue covers the wound bed so completely that you cannot tell how deep the damage goes until debridement reveals the base. “Deep tissue injury” describes intact or disrupted skin with a localized area of purple or maroon discoloration suggesting damage to underlying tissue. In people with lighter skin, this presents as classic purple-maroon discoloration with a defined border. In people with darker skin, clinicians should look for persistent redness and hyperpigmentation rather than blanching, since color changes may not follow the classic presentation.5PubMed Central. Differential diagnosis of suspected deep tissue injury Deep tissue injuries can be confused with Stage 2 ulcers, bruises, incontinence-related skin damage, or even more serious conditions like necrotizing fasciitis, making careful assessment critical.
Diabetic Foot Ulcer Systems
Diabetic foot ulcers sit at the intersection of neuropathy, poor blood supply, and infection, and no single classification has been universally adopted. At least five validated systems exist, each emphasizing slightly different variables.6PubMed Central. Comparison of five systems of classification of diabetic foot ulcers and predictive factors for amputation The two most commonly encountered are the Wagner (sometimes called Meggitt-Wagner) system and the University of Texas (UT) system.
The Wagner system uses a single numerical scale from Grade 0 to Grade 5. Grade 0 is a foot with bony deformities but intact skin, essentially a “foot at risk.” Grade 1 is a superficial ulcer. Grade 2 extends deeper into tissue. Grade 3 involves deep abscess or bone infection. Grade 4 is partial gangrene of the forefoot, and Grade 5 is extensive gangrene requiring urgent intervention.7PubMed Central. Wagner’s Classification as a Tool for Treating Diabetic Foot Ulcers: Our Observations at a Suburban Teaching Hospital Wagner’s appeal is its simplicity: a single number captures the overall severity, making it easy to communicate and remember.
The UT system takes a different approach by adding a second dimension. It grades wound depth along a similar 0-to-3 scale but then stages each grade by the presence or absence of infection and ischemia (poor blood flow). A shallow ulcer that is both infected and ischemic gets a very different stage than a shallow ulcer with neither complication. This two-axis design was built to better predict who ends up needing an amputation, because depth alone does not capture the full picture: a shallow wound with a severely compromised blood supply can be more dangerous than a deeper wound with good circulation.8Diabetes Care. Validation of a Diabetic Wound Classification System: The contribution of depth, infection, and ischemia to risk of amputation Head-to-head comparisons have confirmed that the UT system provides finer prognostic detail than Wagner alone.9Diabetes Care. A Comparison of Two Diabetic Foot Ulcer Classification Systems: The Wagner and the University of Texas wound classification systems
More recently, the SINBAD score has gained traction, especially in resource-limited settings. It stands for Site, Ischemia, Neuropathy, Bacterial infection, Area, and Depth. Each component is scored as either 0 or 1, yielding a total from 0 to 6. A prospective study found that patients with low SINBAD scores had high healing rates and minimal need for amputation, while those with scores of 5 or 6 faced significantly greater risk of major limb loss.10PubMed Central. Clinical Evaluation of Diabetic Foot Ulcers Using the SINBAD Scoring System: A Prospective Study SINBAD’s binary scoring makes it straightforward and quick, which can matter when clinicians vary in experience. In inter-rater reliability testing, the Meggitt-Wagner system showed the highest agreement among multiple clinicians, SINBAD and UT had strong agreement, and the more complex PEDIS system had only moderate agreement.11PubMed. Inter-rater reliability of four validated diabetic foot ulcer classification systems
Reliability is a real concern for any of these tools. When 95 clinicians were asked to grade photographs of diabetic foot ulcers using the Wagner and UT systems, agreement was only moderate for both, suggesting that neither system should be used in isolation without additional clinical information.12PubMed Central. Comparing the Meggitt-Wagner and the University of Texas wound classification systems for diabetic foot ulcers: inter-observer analyses In practice, experienced wound-care teams often combine a classification score with bedside assessment, vascular studies, and imaging rather than relying on a single number.
Venous and Arterial Leg Ulcers
Leg ulcers caused by vascular problems account for a large share of chronic wounds. Most are venous in origin, driven by high pressure in the leg veins, while a smaller fraction result from arterial insufficiency.13PubMed. Appraisal of the information content of the C classes of CEAP clinical classification of chronic venous disorders: a multicenter evaluation of 872 patients Some patients have mixed disease involving both. The dominant classification for venous disorders is CEAP, which stands for Clinical severity, Etiology, Anatomy, and Pathophysiology. It was designed to standardize the description of chronic venous disease from mild spider veins all the way to active ulceration.
The clinical (C) component of CEAP runs from C0 (no visible or palpable signs of venous disease) through C6 (active venous ulcer). C5 denotes a healed venous ulcer. C4 captures skin changes like pigmentation and eczema that signal advancing disease before an ulcer actually forms. CEAP has been called the most comprehensive and widely used classification for chronic venous disorders, and its ascending severity has been validated in multicenter evaluations.14PubMed. Appraisal of the information content of the C classes of CEAP clinical classification of chronic venous disorders: a multicenter evaluation of 872 patients
For arterial ulcers, classification leans less on a formal staging system and more on hemodynamic measurement. The ankle-brachial pressure index, a simple ratio comparing blood pressure at the ankle to blood pressure in the arm, remains the cornerstone for diagnosing ischemic ulcers. A low reading signals that not enough blood is reaching the leg to sustain healthy tissue, and treatment pivots from compression (used for venous ulcers) to revascularization.
Threatened Limbs and the WIfI Framework
When a patient presents with a wound on the lower limb and there is a genuine question about whether the limb can be saved, vascular surgeons increasingly turn to the WIfI classification. Developed by the Society for Vascular Surgery, WIfI scores three factors: Wound severity, Ischemia, and foot Infection. Each factor is graded from 0 (none/minimal) to 3 (severe), and the combination produces a composite risk estimate for amputation within one year and the likelihood that revascularization will be needed.15PubMed. The Society for Vascular Surgery Lower Extremity Threatened Limb Classification System: risk stratification based on wound, ischemia, and foot infection (WIfI)
WIfI was designed to fill a gap left by older systems. Previous classifications for peripheral artery disease, such as the Fontaine and Rutherford scales, focused mainly on symptoms like pain and walking distance, which do not capture wound characteristics or infection. WIfI explicitly combines all three limb-threatening variables, making it more useful for guiding decisions about who needs an urgent bypass or angioplasty versus who can be managed conservatively.16PubMed Central. WIfI classification: the Society for Vascular Surgery lower extremity threatened limb classification system, a literature review Its adoption has grown rapidly in vascular surgery, though like all newer systems it requires further validation in diverse populations.
Stress Ulcers in Critical Care
Stress ulcers are a distinct category that develops in critically ill patients, driven not by chronic disease but by the body’s acute physiological response to severe illness or injury. They form in the stomach or upper small intestine when the normal protective mucosal barrier breaks down under extreme physical stress. Two historically important subtypes carry specific names. Ulcers associated with head injury are called Cushing’s ulcers, while those linked to extensive burns are called Curling’s ulcers.17PubMed. Curling Ulcer These names reflect the conditions that trigger them rather than any difference in the ulcer itself: both are stress-induced erosions, but the underlying mechanisms differ slightly. Cushing’s ulcers are thought to arise from acid hypersecretion triggered by increased intracranial pressure, while Curling’s ulcers relate more to mucosal ischemia from the systemic effects of burns.
Stress ulcers are not formally graded with a numbered staging system the way pressure injuries or peptic ulcers are. Instead, classification in the ICU setting tends to focus on risk stratification: which patients are most likely to develop clinically significant bleeding, and which warrant prophylactic acid suppression. Mechanical ventilation for more than 48 hours and coagulopathy are the two best-established risk factors for stress-related mucosal bleeding.
Atypical Ulcers That Resist Easy Classification
Not every chronic ulcer fits neatly into the venous, arterial, or diabetic categories. A subset of ulcers result from inflammatory, autoimmune, or hypertensive processes that mimic more common wounds but respond poorly to standard care. Recognizing these is crucial because treating them as ordinary ulcers can make them worse.
Pyoderma gangrenosum is a prime example. It is an inflammatory condition in which the skin breaks down into painful, rapidly expanding ulcers, often triggered or worsened by trauma, including surgical incisions. Because it can look like an infected wound, patients sometimes undergo debridement that only aggravates the problem. An international Delphi consensus established formal diagnostic criteria: one major criterion (a biopsy showing a dense infiltrate of neutrophils at the ulcer edge) and eight minor criteria covering clinical features like pathergy, association with inflammatory bowel disease or arthritis, characteristic scarring at healed sites, and improvement with immunosuppressive therapy. Meeting the major criterion plus four of the eight minor criteria yielded a sensitivity of 86% and specificity of 90%.18PubMed. Diagnostic Criteria of Ulcerative Pyoderma Gangrenosum: A Delphi Consensus of International Experts
Martorell’s ulcer, also called hypertensive ischemic ulcer, is another easily misdiagnosed condition. It typically occurs on the lower legs of patients with long-standing, poorly controlled high blood pressure. The mechanism involves thickening and narrowing of small arteries in the skin due to chronic shear stress from elevated pressure, eventually choking off blood flow and causing tissue death.19PubMed Central. Martorell’s Ulcer: Diagnostic and Therapeutic Challenge The ulcers are extremely painful and frequently mistaken for venous ulcers, arterial ulcers, or vasculitis. Biopsy showing characteristic arterial wall changes is the key to diagnosis, and treatment centers on aggressive blood pressure control rather than compression or debridement.
Malignant Transformation in Chronic Wounds
A chronic wound that refuses to heal over many years can occasionally undergo malignant transformation, a phenomenon known as a Marjolin ulcer. These are typically squamous cell carcinomas arising in burn scars, chronic venous ulcers, or other long-standing wounds, sometimes decades after the original injury. Standard cancer staging systems were not designed with these ulcers in mind, and a single-center study found that the AJCC staging system classified the vast majority of Marjolin tumors as a single stage (T3 in about 71% of cases), offering no meaningful ability to distinguish which patients would experience recurrence or death.20Burns. Marjolin ulcer: A single-center experience and evaluation of staging systems When researchers applied exploratory depth thresholds instead, the results were far more telling: patients with shallow invasion had no distant spread, while those with deep invasion had recurrence rates of 80% and distant metastasis in 60% of cases. The strongest independent predictor of poor outcome was whether cancer had already spread to lymph nodes.
The clinical implication is that any chronic ulcer that has been present for months to years without healing, particularly one that changes in appearance, develops raised borders, or starts bleeding in new ways, should be biopsied. Wound-care teams familiar with classification systems for routine ulcers may not always think of malignancy, but it belongs on the differential diagnosis for any wound that defies expectations.
Oral Ulcers and Mucosal Classification
Ulcer classification extends beyond the skin and gastrointestinal tract. Recurrent aphthous stomatitis, the most common cause of recurring mouth ulcers, is divided into three types. Minor aphthous ulcers are small, shallow, and heal within a couple of weeks. Major aphthous ulcers are larger, deeper, and can take weeks to months to resolve, sometimes leaving scars. Herpetiform aphthous ulcers appear as clusters of tiny sores that may merge into larger irregular ulcers.21PubMed. Recurrent aphthous stomatitis – Etiology, serum autoantibodies, anemia, hematinic deficiencies, and management The name “herpetiform” refers to their clustered appearance, not to herpes virus infection. This three-part classification guides how aggressively clinicians investigate underlying causes: minor ulcers in an otherwise healthy person rarely warrant extensive workup, whereas recurrent major ulcers might prompt testing for nutritional deficiencies, celiac disease, or immune disorders.
Corneal ulcers are another mucosal category with their own classification logic. They are broadly divided by cause into infectious (bacterial, viral, fungal, or parasitic) and non-infectious (from dry eye, autoimmune disease, or trauma).22TNOA Journal of Ophthalmic Science and Research. Treatment Approach to Corneal Ulcer Identifying the etiology drives treatment entirely: a bacterial corneal ulcer demands antibiotics, a fungal one needs antifungals, and an autoimmune ulcer may require immunosuppression. Misclassification in the eye can lead to vision loss, making corneal ulcer classification one of the higher-stakes examples.
Automated Wound Classification With Deep Learning
One of the persistent problems with ulcer classification is inter-rater variability. As the diabetic foot ulcer studies showed, even experienced clinicians only moderately agree on what grade a wound deserves when looking at the same photographs. This gap has motivated research into automated classification using machine learning. A deep-learning model trained to classify wound images into four distinct types achieved an accuracy of about 92% and an F1 score of around 0.91, demonstrating solid performance across different wound distributions.23arXiv. Deep Learning for Automated Wound Classification And Segmentation The same model performed well on wound segmentation, correctly outlining wound boundaries with a precision of about 95%.
These tools are not replacing clinical judgment yet, but they point toward a future where a smartphone image could provide an initial wound classification in a rural clinic or a home-care visit. The challenge is that current models are trained on limited datasets and may not generalize well across skin tones, wound types not represented in the training data, or real-world image quality. Still, automated classification could reduce the variability problem that plagues every human-scored system described in this article, potentially catching high-risk wounds earlier.