Hematoxylin and eosin, almost always abbreviated H&E, is the most widely used staining method in medical pathology and has been for well over a century. When a surgeon removes a suspicious lump or a dermatologist punches a tiny core from a mole, the tissue nearly always ends up on a glass slide stained with these two dyes so a pathologist can examine it under a microscope. Hematoxylin colors cell nuclei a deep blue-purple, while eosin turns proteins in the cytoplasm and surrounding tissue varying shades of pink. That simple color contrast reveals an enormous amount about what is happening inside the body, from routine inflammation to aggressive cancer.
A Dye From the Rainforest
Hematoxylin is not a synthetic chemical cooked up in a factory. It comes from the heartwood of a tropical tree called Haematoxylum campechianum, native to Central America and the Yucatán Peninsula. Spanish explorers encountered the tree in 1502 and quickly recognized the commercial value of its deep red-purple extract. A thriving trade developed around growing, harvesting, and shipping the wood to Europe, where it was used to dye fabrics long before anyone thought to put it on a tissue slide. That pivot happened in the mid-1800s, when amateur microscopists discovered that hematoxylin could color cellular structures in ways nothing else could at the time.1PubMed. The long history of hematoxylin
Eosin, by contrast, is a synthetic fluorescent dye first produced in the late 19th century. Pairing the two was a stroke of practical genius: one dye that highlights nuclei and another that highlights everything else gives pathologists a quick, reliable map of any tissue’s cellular architecture. The combination stuck because it works on nearly every tissue type, is cheap to prepare, and produces consistent results across laboratories worldwide.
From Tissue Sample to Glass Slide
Before H&E staining can happen, the tissue has to be preserved and sliced thin enough for light to pass through. The standard route starts with formalin fixation: the fresh tissue is soaked in a buffered formalin solution, which cross-links proteins and halts decay, essentially freezing the tissue in its current biological state.2PubMed. Making Formalin-Fixed, Paraffin Embedded Blocks Once fixed, the tissue is dehydrated through a series of alcohol baths, then infiltrated with molten paraffin wax. The wax solidifies around the tissue, creating a firm block that can be mounted on a microtome, a precision cutting instrument that shaves sections roughly four to five micrometers thick.3PubMed. Histopathology procedures: from tissue sampling to histopathological evaluation Those wafer-thin slices are floated onto glass slides, the paraffin is dissolved away, and the tissue is rehydrated before the dyes are applied.
The whole process, from fresh biopsy to stained slide, typically takes about a day in a routine clinical lab. Most of that time is consumed by fixation and processing rather than the staining itself, which can be done in minutes. Speed matters because patients and surgeons are waiting for answers, and delays in tissue processing are a persistent bottleneck in busy hospitals.
How the Two Dyes Actually Work
Hematoxylin on its own does not stain anything useful. It first has to be oxidized into a related compound called hematein, which then pairs with a metal ion, usually aluminum, to form what chemists call a “lake.” That metal-dye complex carries a positive charge and is drawn to negatively charged molecules in the cell, particularly the phosphate groups in DNA and RNA packed inside the nucleus. The result is a vivid blue-purple nuclear stain.4PubMed. Nuclear staining with alum hematoxylin
Eosin works by a complementary mechanism. It is an acidic dye carrying a negative charge, so it binds to positively charged amino acid residues in proteins, particularly arginine, histidine, and lysine.5PubMed. Mechanism of dye binding in the protein assay using eosin dyes Because proteins fill the cytoplasm, the connective tissue between cells, collagen fibers, and red blood cells, eosin effectively stains everything the hematoxylin does not. The pink-to-red spectrum of eosin varies with protein density: muscle fibers stain a darker pink, loose connective tissue a paler shade, and red blood cells an intense cherry red.
Together, the two dyes create a high-contrast image where every cell’s nucleus stands out sharply against its surrounding cytoplasm. Researchers have noted that the precise chemistry behind how the hematoxylin-metal complex binds to nucleic acids is still not fully understood, despite more than 150 years of use.6PubMed. Hematoxylin and eosin staining of tissue and cell sections It is one of those cases where a technique became indispensable well before scientists could completely explain its mechanism.
What Pathologists See and Why It Matters
An H&E-stained slide gives a pathologist a surprising amount of diagnostic information. Nuclei show varying patterns of chromatin condensation, the way DNA is packaged inside the nucleus, and those patterns differ between cell types and between normal and cancerous cells.7PubMed. Hematoxylin and eosin staining of tissue and cell sections A pathologist can often distinguish a benign growth from a malignant one just by looking at how dark, how large, and how irregular the nuclei appear, and by checking whether the tissue’s normal architecture has been disrupted.
In breast cancer, for instance, pathologists use the architectural features visible on H&E slides to assign a tumor grade, which predicts how aggressively the cancer is likely to behave.8PubMed. Computer-aided prognosis on breast cancer with hematoxylin and eosin histopathology images: A review They assess how much the tumor cells resemble normal breast tissue, how often the cells are dividing, and how varied the nuclear shapes are. All of those features are visible on a standard H&E slide without any additional molecular testing.
Beyond cancer, H&E staining reveals inflammation (clusters of immune cells flooding into tissue), infection (the architectural damage and cellular debris it causes), fibrosis (excess collagen replacing normal tissue), and a long list of other conditions. A single H&E slide can be the difference between a diagnosis of celiac disease and irritable bowel syndrome, or between a benign nevus and melanoma. It is the first test ordered on almost every surgical specimen and the foundation on which more specialized tests are built.
When H&E Is Not Enough
For all its versatility, H&E has real blind spots. It stains based on broad chemical properties, positive versus negative charge, not on the identity of specific molecules. That means certain organisms, deposits, and tissue components can be difficult or impossible to pick out.
A good example is the bacterium Helicobacter pylori, which infects the stomach lining and can lead to ulcers and cancer. One comparative study found that H&E had a sensitivity of about 67% for detecting the organism, meaning it missed roughly a third of confirmed cases. A modified Giemsa stain performed better at around 87% sensitivity, and immunohistochemistry, which uses antibodies to target the bacterium specifically, was the most sensitive of all, especially when only a few organisms were present.9GLOBAL JOURNAL FOR RESEARCH ANALYSIS. A COMPARATIVE STUDY OF HEMATOXYLIN & EOSIN, SPECIAL STAINS AND IMMUNOHISTOCHEMISTRY FOR DETECTION OF HELICOBACTER PYLORI IN CASES OF GASTRITIS
Another tricky situation is detecting amyloid, an abnormal protein deposit that can accumulate in skin and organs. In one study of patients with cutaneous amyloidosis, H&E revealed the deposits in only about 36% of confirmed cases. Crystal violet stain picked up the same percentage, but neither matched the performance of more targeted methods.10PubMed Central. Comparison of Immunostaining with Hematoxylin-Eosin and Special Stains in the Diagnosis of Cutaneous Macular Amyloidosis These limitations do not make H&E a poor stain; they mean it is a generalist. When a pathologist suspects something specific, they reach for additional tools.
That said, H&E outperforms specialized stains in some contexts. A study comparing several approaches for detecting keratin in oral tissues found that H&E had the highest staining specificity and intensity compared to special stains designed for the same purpose.11PubMed Central. Comparison of Special Stains for Keratin with Routine Hematoxylin and Eosin Stain So the question is never “is H&E good or bad?” but rather “what is H&E good at, and where does something else do a better job?”
Frozen Sections During Surgery
One of the most dramatic uses of H&E happens while a patient is still on the operating table. During surgery, a pathologist may receive a fresh tissue sample for what is called a frozen section analysis. Instead of the day-long formalin-fixation-and-paraffin process, the tissue is rapidly frozen, sliced on a specialized microtome inside a cryostat, and stained with a quick version of H&E. The pathologist examines the slide and relays results to the surgeon within minutes, often informing whether a tumor margin is clear or whether more tissue needs to come out.
Frozen sections are indispensable in breast surgery, where the decision about how much tissue to remove can hinge on intraoperative pathology.12PubMed Central. Comparison of staining quality between rapid and routine hematoxylin and eosin staining of frozen breast tissue sections: an observational study However, the technique has limitations. The rapid H&E mixture is unstable and has to be freshly prepared right before use, which adds workload for lab technicians. Frozen sections also tend to have more ice-crystal artifacts and slightly worse morphological detail compared to permanent paraffin sections. In poorly differentiated tumors, where cells look especially abnormal and hard to classify, conventional H&E staining on frozen sections sometimes cannot provide a definitive answer, and additional techniques like rapid immunohistochemistry may be needed.13APMIS. Rapid Automated Immunohistochemistry on Frozen Sections Enables Real-Time Surgical Pathology Decisions
Artifacts and Quality Control Headaches
Anyone who has worked in a histology lab knows that a beautiful H&E slide is not guaranteed. Things can go wrong at every step, from the moment the surgeon cuts the tissue to the final mounting of the coverslip. Artifacts, features on the slide that do not represent real tissue structures, can creep in during fixation (tissue shrinkage, incomplete preservation), processing (air bubbles, uneven wax infiltration), microtomy (knife marks, folds, chattering), and staining itself (uneven dye uptake, precipitate deposits). Some artifacts merely annoy the pathologist; others can mimic disease or obscure it entirely, rendering the slide useless.14PubMed Central. A review of artifacts in histopathology
Staining variability is a particularly stubborn issue. Even slides prepared from the same tissue block can look noticeably different if they are stained on different days, with different batches of dye, or by different technicians. In routine diagnostic work this rarely causes problems because experienced pathologists mentally compensate for slight color shifts. But as we will see, the problem becomes much more serious when computers enter the picture.
AI and the Stain Variation Problem
The push to apply artificial intelligence to pathology has put H&E slides at the center of a rapidly growing field. Researchers have trained deep learning algorithms to analyze whole-slide images of H&E-stained tumors and predict outcomes like overall survival, sometimes matching the diagnostic accuracy of experienced pathologists.15npj Precision Oncology. Usability of deep learning and H&E images predict disease outcome-emerging tool to optimize clinical trials In one study on bladder cancer, an AI system’s diagnostic performance matched that of 17 pathologists when classifying urothelial carcinoma on whole-slide images.16Nature Machine Intelligence. Pathologist-level interpretable whole-slide cancer diagnosis with deep learning
These results are exciting, but they come with a major caveat: stain variation. A deep neural network trained on one batch of H&E slides can fail catastrophically when tested on a different batch prepared from the same tissue blocks at a different time. One study found that a model performing well on its training batch (area under the curve of 0.74 to 0.81) dropped to essentially random guessing when applied to slides stained from the same blocks in a separate batch (area under the curve of 0.52 to 0.53). Even applying color normalization techniques, including sophisticated adversarial network approaches, did not rescue the performance.17Scientific Reports. Impact of stain variation and color normalization for prognostic predictions in pathology
This is a problem unique to computational pathology. Human pathologists handle stain variation intuitively because they are reading patterns and relationships, not pixel colors. AI models, especially those trained on images from a single institution, can latch onto batch-specific color profiles rather than the tissue features that actually matter. Until this problem is solved, deploying AI diagnostic tools across different hospitals with different staining protocols remains a significant challenge.
The Xylene Problem
A less glamorous but very real concern in H&E work is the chemical used to clear the tissue during processing. Xylene is the traditional clearing agent, the solvent that makes paraffin-infiltrated tissue transparent before staining. It works beautifully, but it is also toxic. Chronic exposure causes headaches, dizziness, and respiratory irritation, and prolonged contact can damage the liver and kidneys. Lab technicians who handle xylene daily are the most affected, and awareness of these health risks has driven a steady search for safer alternatives.18PubMed. Safer and Sustainable Alternatives to Xylene in Histopathology: A Comprehensive Review
Several substitutes have been tested. A commercial product called UltraClear was found to produce tissue sections that cut well and stained properly with both H&E and immunohistochemistry.19PubMed Central. Alternative to xylene as a clearing agent in histopathology Another group developed a mixture of white oil and n-heptane, designated SBO, which had a high boiling point, low volatility, and no odor, while still producing H&E slides with clear nuclear and cytoplasmic detail comparable to xylene-processed tissue.20PubMed Central. A novel non-toxic xylene substitute (SBO) for histology Despite these promising results, adoption has been slow. Many labs remain hesitant to change a workflow that has been validated over decades, and regulatory requirements for demonstrating equivalent diagnostic quality create a high bar for any new reagent.
A Fragile Supply Chain
Because hematoxylin comes from a tree rather than a chemical synthesis plant, its supply is inherently vulnerable. The chain from forest to laboratory involves multiple separate entities: companies that grow and harvest the trees, others that extract the dye from the wood, still others that purify it into usable hematoxylin, and finally the vendors that formulate and sell staining solutions to labs. A breakdown at any link, a poor harvest, a processing plant closure, a trade disruption, can trigger shortages that ripple through clinical laboratories worldwide.21PubMed. Hematoxylin shortages: their causes and duration, and other dyes that can replace hemalum in routine hematoxylin and eosin staining
These shortages have happened repeatedly over the decades, prompting pathologists and histotechnologists to investigate synthetic nuclear stains that could replace hematoxylin if supplies dried up entirely. Celestine blue, gallocyanin, and nuclear fast red are among the alternatives that have been explored. None have displaced hematoxylin in routine use, largely because pathologists have spent their entire careers learning what normal and abnormal tissues look like in the specific blue-purple palette that hematoxylin provides. Switching nuclear stains would mean retraining countless professionals to interpret a different color map, a far larger obstacle than chemistry alone.
H&E in Forensic Wound Dating
Outside routine clinical diagnosis, H&E plays a specialized role in forensic pathology, particularly in estimating when a wound was inflicted. The cellular changes that follow an injury unfold on a predictable timeline: early on, you see hemorrhage and infiltration by certain immune cells; later, fibroblasts appear and begin laying down new collagen. H&E staining can visualize these sequential changes and help forensic pathologists place a wound in a timeframe.
In autopsy-based research on human skin abrasions, the histological changes visible on H&E, including the shift from early inflammatory cell infiltration to fibroblast proliferation and collagen deposition, showed a significant correlation with wound age beyond about 72 hours.22PubMed. Predictive accuracy of histopathological profile and immunohistochemical markers for the aging of abrasion: an autopsy-based study In veterinary forensic work, H&E staining proved useful for dating wounds in their early phase, less than about five days old, by detecting newly formed collagen fibers and signs of red blood cell breakdown.23PubMed. Forensic wound age estimation in dog tissue correlated with newly formed collagen fibres: a retrospective study This kind of evidence can be critical in criminal investigations, where establishing whether injuries occurred before or after death, or days versus hours before discovery, has direct legal significance.
Wound dating illustrates something broader about H&E: its value lies not just in identifying what is present in a tissue sample but in reading the biological story the tissue tells over time. The same fundamental stain that helps a breast surgeon decide how much tissue to remove can help a forensic pathologist reconstruct what happened to a victim. That versatility, built on the simple contrast between a blue nuclear dye and a pink protein dye, is why H&E has outlasted every attempt to replace it.