Imagine turning a bottle around at the table.
The front label offers a hillside, a family name, perhaps a few words about the harvest. The back offers something much less romantic:
Contains sulfites.
For someone with a diagnosed sensitivity, those words matter immediately. For someone else, they bring back a headache and a question that was never properly answered. A third person hears “preservative” and wonders whether the wine has been made less honest.
The same statement has become a warning, an explanation and a judgment about quality.
It cannot do all three jobs by itself.
In the previous Cellar Journal article, we followed what skin contact can bring into a wine. Now we turn to the next decision: how to protect what the producer has chosen to make—and how to discuss that protection without dismissing the person who will drink it.
Sulfites are a tool with advantages and limits. They do not explain every headache, and they do not replace careful winemaking.
Understanding their role does not require defending every addition. Taking a reaction seriously does not require pretending that its cause is already known.
What are sulfites in wine?
In this context, sulfites is an umbrella term covering sulfur dioxide and sulfiting agents. It is not another name for every substance containing sulfur. (TTB: declaration of sulfites).
Sulfur dioxide can enter the story through fermentation as well as deliberate addition. Research with selected wine-yeast strains measured its production during white-wine fermentation. That does not establish one concentration for all wines; it establishes why a decision not to add sulfites is different from a result showing none present. (Eglinton & Henschke, 1996).
A winery’s addition record and a laboratory’s analysis answer different questions.
“No added sulfites” describes a production choice. It is not an analytical zero.
That distinction gives the claim more precision, not less value. A producer can sincerely choose an approach with no additions, and a consumer can prefer it. Neither needs to turn the preference into a chemical claim the bottle has not demonstrated.
What the preservative is trying to preserve
Sulfur dioxide helps limit unwanted microbial activity and oxidative changes. Iowa State University’s winemaking guidance places it alongside sanitation and other protective practices—not in place of them. (Iowa State University: sulfur dioxide in winemaking).
The aim may be to preserve fruit, avoid premature browning or reduce the opportunity for unwanted organisms to alter the wine. Calling that protection “antioxidant” describes a role in the liquid. It is not a health claim for the person drinking it.
Nor does the chemistry work like a tiny sponge that simply removes every molecule of oxygen. In model-wine experiments, sulfur dioxide intercepted hydrogen peroxide involved in oxidation chemistry, limiting a pathway that can produce acetaldehyde from ethanol. The mechanism is more specific than “oxygen goes in, preservative cancels it.” (Elias & Waterhouse, 2010).
The practical question is therefore not whether preservation belongs in wine. Every producer manages change in some way.
It is which changes are being encouraged, which are being limited, and how successfully the decisions preserve the intended style.
A preservative cannot retrospectively turn damaged fruit into healthy fruit or make an inaccurate label truthful. Its usefulness is not a certificate for everything that happened before it was used.
Free, bound and total: three readings, not three ingredients
A technical sheet may give a sulfur-dioxide number without saying which number it is.
That missing word matters.
Free SO₂ includes forms in chemical equilibrium, among them molecular sulfur dioxide and bisulfite. Bound SO₂ is associated with other wine components. Total SO₂ combines the free and bound fractions. Molecular SO₂ is part of the free fraction, not a fourth quantity to add to the total. (Iowa State University: measuring SO₂).
The OIV’s molecular-SO₂ method uses pH, alcohol and temperature to calculate the molecular fraction. At otherwise comparable conditions, lower pH increases that fraction. This is why identical total-SO₂ values do not establish identical antimicrobial protection. (OIV molecular-SO₂ method).
Here is an intentionally incomplete comparison. Two technical sheets report the same total, but neither gives free SO₂, pH or the date of analysis. It would be tempting to call the wines equally protected.
The information does not support that conclusion.
The comparison is missing the very conditions that help interpret the number. A concentration without its measurement type and context can look more informative than it is.
There is a further nuance: bound does not mean permanently irrelevant. A 2022 NMR study observed dissociation of acetaldehyde-bound sulfur dioxide under acidic, wine-like conditions, with implications for oxidative stability. That finding does not make every bound form equally available; it warns against treating all binding as irreversible. (Tachtalidou et al., 2022).
For a reader, there is no need to memorize an equilibrium equation. Remember the distinction:
Total helps describe how much is measured. Free and molecular fractions help interpret how part of it may function. Neither is a personal medical clearance.
Protection continues after the cellar decision
A sulfur-dioxide addition is one event. A bottle’s life is a sequence.
The AWRI separates oxygen dissolved in wine from oxygen in the package headspace and contributions associated with the closure. These can affect the wine after filling. A bottling number is not a promise that the protective situation remains unchanged indefinitely. (AWRI: total package oxygen).
The institute’s technical guidance also describes interactions among pH, sulfur-dioxide management and microbial or oxidation problems. Its practical message is to interpret the system, not simply celebrate the smallest addition. (AWRI: revisiting sulfur dioxide use).
This is where the natural-wine conversation becomes more useful. “How little was used?” may be a legitimate question. “How was the wine protected?” is a broader one.
The answer might involve handling, monitoring and packaging as well as additions. Less visible intervention does not necessarily mean less work. More preservative does not automatically mean better work.
Can you smell sulfites?
Sulfur dioxide can produce a sharp, pungent impression when sufficiently apparent. The familiar rotten-egg description belongs instead to hydrogen sulfide, a different sulfur compound. The AWRI treats them separately in its account of wine faults. (AWRI: sulfur dioxide and hydrogen sulfide).
This distinction is small in spelling and large in meaning.
A note saying “sulfur” is not a laboratory report. It cannot quantify total sulfites, reveal an addition history or establish whether a wine is suitable for someone with sensitivity. Conversely, not noticing a pungent smell cannot establish that none are present.
Use sensory language to describe the impression. Use appropriate evidence for the chemical or medical claim.
What “contains sulfites” actually tells you
For wines covered by the U.S. TTB labeling guidance, a declaration is required at 10 or more parts per million, measured as total sulfur dioxide. The statement alerts consumers to presence; it does not give the amount added or a complete production history. (TTB: declaration of sulfites).
The European Union’s food-information regulation uses more than 10 mg/kg or 10 mg/litre, expressed as total SO₂, for the listed sulfur-dioxide/sulfite category. For a beverage, the relevant concentration is expressed per litre. The inequality is not worded identically to the U.S. guidance. (EU food-information regulation, Annex II).
Those are disclosure thresholds within particular rules. They are not interchangeable with a maximum permitted addition, a sensory threshold or a personal reaction threshold.
| Wording or evidence | The question it helps answer | What remains unanswered |
|---|---|---|
| A mandatory declaration | Has the applicable disclosure condition been met? | How much was deliberately added; why someone developed symptoms |
| “No added sulfites” | What addition practice is claimed? | Whether fermentation produced sulfites |
| A dated total-SO₂ result | What total was measured in that sample? | Its free fraction, complete protection or medical suitability |
| An organic claim | Which certification rules apply? | Whether the bottle is symptom-free for a particular person |
TTB does not permit absolute statements such as “sulfite free”; it distinguishes permitted analytical wording, including “no detectable sulfites” under its testing conditions. Absence of a declaration should not be read as a universal proof of zero. (TTB: declaration of sulfites).
The useful habit is not to become a label inspector over dinner. It is to stop asking a short statement to supply information it was never designed to provide.
Organic does not settle every sulfite question
In the United States, USDA distinguishes organic wine, which cannot contain added sulfites, from wine made with organic grapes, which may contain them under the applicable conditions and cannot use the USDA organic seal. These are different regulated claims. (USDA: organic alcohol categories).
Read that example with its jurisdiction attached. It is not a worldwide rule that every bottle described as organic must follow the U.S. category.
Nor does a farming or processing commitment tell us how your body will respond. Choosing a verified production approach is one decision; evaluating a recurrent reaction is another. A label may help with the first without being able to settle the second.
Sulfite sensitivity is real—and it deserves the right question
A responsible discussion must not turn skepticism about a popular explanation into skepticism about people’s symptoms.
ASCIA, the Australasian Society of Clinical Immunology and Allergy, identifies respiratory reactions as an important concern, particularly in people with asthma. Wheezing, cough or chest tightness after exposure deserve attention. Not every sulfite reaction follows the mechanism of a conventional allergy. (ASCIA: sulfite sensitivity).
Clinical research also shows why stories about “wine sensitivity” need careful evaluation. In a controlled study of people reporting wine-induced asthma, only a subset reacted under the tested sulfite-challenge conditions. The result supports a role for sulfites in some reactions, not a universal explanation for everyone who reports symptoms with wine. (Vally & Thompson, 2001).
Two statements can therefore be true:
A reaction may be genuine. Its cause may still require investigation.
For recurring symptoms, discuss the pattern with a qualified clinician rather than testing a succession of supposedly safer bottles. ASCIA notes that routine blood or skin tests do not reliably establish sulfite sensitivity; a specialist may consider a supervised challenge when appropriate. This is not a home-tasting exercise. (ASCIA: sulfite sensitivity).
Do sulfites cause wine headaches?
The honest answer is not “every time.” It is also not “never.”
A headache soon after wine, a migraine episode and the headache within a next-day hangover should not automatically be treated as the same event. Mayo Clinic describes a wine-related headache occurring within hours as distinct from a hangover and notes uncertainty about its cause. (Mayo Clinic: wine headache and hangover).
NIAAA identifies alcohol as the principal contributor to hangovers, while also acknowledging that sulfites may contribute to headaches in sensitive people. Reducing one component does not remove ethanol or its other effects. (NIAAA: hangovers).
The evidence is not one-directional. A 2019 study of 80 volunteers aged 18–25 reported an association between higher sulfite exposure and headaches in a susceptible subgroup. Yet a separate paired comparison in the same paper found no significant difference between the lower- and higher-sulfite conditions. Its questionnaire-based outcomes and narrow population call for caution, not a universal verdict. (Silva et al., 2019).
Other explanations remain under investigation. In 2023, researchers reported that a quercetin metabolite inhibited an enzyme involved in acetaldehyde metabolism and proposed a mechanism for some red-wine headaches. This was an enzyme-based hypothesis, not a clinical demonstration in drinkers; the authors explicitly called for human testing. It should not be repackaged as the discovery of the one true cause. (Devi, Levin & Waterhouse, 2023).
The point is not to replace one suspect with another fashionable molecule.
It is to keep the symptom, the timing, the exposure and the strength of the evidence separate. A headache that follows wine is information worth recording. By itself, it is not an ingredient analysis.
What food-safety assessments add—and what they do not
Individual sensitivity is not the only question regulators study.
In 2022, EFSA reported data gaps and potential concern for high dietary consumers, withdrawing its temporary acceptable daily intake because the evidence was insufficient to establish an ADI. That is a population-level food-additive assessment, not a diagnosis of a particular headache. (EFSA: 2022 assessment and data gaps).
A 2025 EFSA report modeled dietary exposure under alternative maximum-use levels. A modeled scenario is not an enacted limit, and a dietary assessment is not a calculation telling an individual how much wine will be symptom-free. (EFSA: 2025 dietary-exposure update).
These findings deserve neither dismissal nor conversion into a dramatic number of “safe glasses.” Different questions require different evidence.
What to do with a reaction—not with a home chemistry kit
If symptoms recur, retain the label, producer and vintage details and write down the timing and symptoms you actually experienced. That record can support a medical conversation. Do not deliberately reproduce a reaction to fill the notebook.
Difficulty breathing or swallowing, swelling of the lips or throat, or collapse calls for urgent medical help, not continued tasting or a wait for the wine to “open up.” Contact emergency services. (Cleveland Clinic: symptoms and urgent care).
A consumer treatment that changes a chemical measurement is not, by that fact alone, a clinical safety test. We do not recommend using drops, aeration or a home treatment to make a suspect wine suitable for a diagnosed sensitivity. Nor does a softer smell establish safety.
That is a boundary on what the evidence reviewed here can support, not a claim that every treatment works identically or that every commercial product has been tested.
The Five-Layer Sulfite Read
This is an original Ask Sommelier AI editorial framework for understanding claims. It is not a dosage calculator, allergy test or instruction to challenge your tolerance.
1. Label claim
Start with the exact words: contains, no added, organic, or a stated concentration. Identify the market and any applicable certification. Do not quietly replace the words with the claim you hoped to read.
2. Total and free SO₂
When an analysis is available, ask what was measured, for which wine or batch, by which method and on what date. If only total is reported, keep free unknown. Missing data should stay missing.
3. pH and wine matrix
Ask whether there is enough context to interpret the analysis. The same printed total is not a complete account of protection. Do not infer laboratory pH from how sharp a sip seems.
4. Oxygen and microbial risk
Look for verified information about handling and packaging rather than assuming that addition level alone describes the producer’s care. This layer concerns the wine’s condition, not the drinker’s medical tolerance.
5. Sensory and medical boundary
Separate three statements: I notice a pungent aroma; a laboratory measured this concentration; I experienced these symptoms. They are different observations, and each deserves the appropriate next question.
The framework need not become a questionnaire delivered to a server. For an ordinary production preference, one question may be enough:
“When you say no added sulfites, is that a verified statement for this wine and vintage?”
For a known sensitivity, a reassuring description is not enough. Make the medical requirement explicit and follow the guidance of your clinician rather than relying on a style category.
A better role for Ask Sommelier AI
A recommendation system should preserve exact label wording, source, vintage, jurisdiction and available analysis. It should keep not added separate from not detected, and a stylistic preference separate from a medical restriction.
It must not transform “natural,” “organic,” “white,” “red,” or “low sulfite” into a promise that someone will not react. When bottle-specific evidence is absent, uncertainty should remain visible.
These are standards for responsible assistance, not a claim that an app can replace clinical evaluation.
Return to the bottle at the table.
The front label still tells a story. The back label still matters. Neither needs to become an argument for or against the person holding the glass.
Protecting the wine and taking the drinker seriously are not opposing commitments.
The preservative deserves an accurate explanation. A reported reaction deserves care. Both become harder to understand when one word is made responsible for everything.
Next, we will ask what happens when a bottle’s unexpected character reaches the glass: is it a fault, a style, or simply something unfamiliar?
Frequently asked questions
Does “no added sulfites” mean none are present?
No. Fermentation can contribute sulfites. TTB even provides wording that distinguishes no additions from naturally occurring sulfites. An addition claim and an analytical result are different evidence. (TTB: naturally occurring sulfites and label wording).
Does red wine always contain fewer sulfites than white?
ASCIA describes a general tendency toward higher concentrations in white wine, not a guarantee for every bottle. Color cannot establish a concentration or make a wine suitable for a sensitive person. (ASCIA: sulfite sensitivity).
Is sulfite sensitivity the same as an allergy to sulfonamide antibiotics?
No. ASCIA distinguishes these conditions; sulfites, sulfates and sulfonamide medicines are not interchangeable categories. Describe the actual substance and reaction to a clinician rather than using the broad phrase “sulfur allergy.” Do not change prescribed treatment on the basis of this article. (ASCIA: sulfonamide allergy is a different question).
Can a laboratory’s total-SO₂ value tell me whether I will get a headache?
No. It describes the analyzed sample, not an individual clinical outcome. The headache studies discussed above do not supply a validated bottle-by-bottle prediction for a particular reader. (Silva et al., 2019; Devi, Levin & Waterhouse, 2023).
Can decanting or sulfite-removal drops certify a wine as safe for me?
No such safety guarantee follows from the evidence reviewed here. Do not use a home treatment or deliberate re-exposure to test suspected sensitivity. Medical assessment and sensory experimentation are different activities. (Cleveland Clinic: symptoms and urgent care).
Is this article medical advice or a cellar dosing guide?
Neither. It explains wine practices, label distinctions and evidence boundaries. Recurrent reactions require individual medical assessment; cellar additions require competent analysis and the applicable production rules.
References and further reading
Links identify the underlying paper or official guidance. Access levels and methodological limitations are documented in the companion research dossier.
- Eglinton, J. M., & Henschke, P. A. (1996). Saccharomyces cerevisiae strains AWRI 838, Lalvin EC1118 and Maurivin PDM do not produce excessive sulfur dioxide in white wine fermentations. Australian Journal of Grape and Wine Research, 2, 77–83. Source / DOI.
- Iowa State University Extension and Outreach (2022). Sulfur Dioxide in Winemaking. FS 0053; April 2022. Source / DOI.
- Elias, R. J., & Waterhouse, A. L. (2010). Controlling the Fenton reaction in wine. Journal of Agricultural and Food Chemistry, 58(3), 1699–1707. Source / DOI.
- International Organisation of Vine and Wine (OIV) (accessed 2026-09-12). Sulfur dioxide (molecular method) (Type-IV). OIV-MA-AS323-04C. Source / DOI.
- Savits, J. / Iowa State University Extension and Outreach (2019). Sulfur Dioxide Measurement and Overestimation in Red Wine. Institutional technical article, April 26. Source / DOI.
- Tachtalidou, S., et al. (2022). Direct NMR evidence for the dissociation of sulfur-dioxide-bound acetaldehyde under acidic conditions: Impact on wines oxidative stability. Food Chemistry, 373, Part B, 131679; online November 2021. Source / DOI.
- Australian Wine Research Institute (AWRI) (accessed 2026-09-12). Revisiting sulfur dioxide use. Institutional technical guidance. Source / DOI.
- Australian Wine Research Institute (AWRI) (accessed 2026-09-12). Oxygen pick-up during packaging — understanding total package oxygen. Institutional technical guidance. Source / DOI.
- Australian Wine Research Institute (AWRI) (accessed 2026-09-12). Wine flavours, faults and taints. Sections on sulfur dioxide and hydrogen sulfide. Source / DOI.
- Alcohol and Tobacco Tax and Trade Bureau (TTB) (accessed 2026-09-12). Wine Labeling: Declaration of Sulfites. Official U.S. guidance; TTB G 2019-3. Source / DOI.
- European Parliament and Council / EUR-Lex (consolidated 2025-04-01; accessed 2026-09-12). Regulation (EU) No 1169/2011 on the provision of food information to consumers. Annex II, point 12; consolidated reference 02011R1169-20250401. Source / DOI.
- USDA Agricultural Marketing Service (accessed 2026-09-12). Organic Alcohol. Official certification guidance. Source / DOI.
- Australasian Society of Clinical Immunology and Allergy (ASCIA) (2024; accessed 2026-09-12). Sulfite Sensitivity. Patient guidance updated August 2024. Source / DOI.
- Vally, H., & Thompson, P. J. (2001). Role of sulfite additives in wine induced asthma: single dose and cumulative dose studies. Thorax, 56(10), 763–769. Source / DOI.
- Silva, M., et al. (2019). Sulfite concentration and the occurrence of headache in young adults: a prospective study. European Journal of Clinical Nutrition, 73, 1316–1322. Source / DOI. Author-copy text consulted.
- National Institute on Alcohol Abuse and Alcoholism (NIAAA) (accessed 2026-09-12). Hangovers. Official health information. Source / DOI.
- Mayo Clinic (accessed 2026-09-12). Hangovers — Symptoms and causes. Clinical patient education. Source / DOI.
- Devi, A., Levin, M., & Waterhouse, A. L. (2023). Inhibition of ALDH2 by quercetin glucuronide suggests a new hypothesis to explain red wine headaches. Scientific Reports, 13, 19503. Source / DOI.
- European Food Safety Authority (EFSA) (2022). Sulfites: safety concern for high consumers, but data lacking. Agency explanation of its reassessment, November 24. Source / DOI.
- European Food Safety Authority; Horváth, Z., & Dujardin, B. (2025). Update of the dietary exposure to sulfur dioxide (E 220) … with alternative maximum levels when used as food additives. EFSA Supporting Publications, 22(11), EN-9754; title abbreviated; published November 19. Source / DOI.
- Cleveland Clinic (2024; accessed 2026-09-12). Sulfite Sensitivity. Medically reviewed patient information, April 24, 2024. Source / DOI.
- Australasian Society of Clinical Immunology and Allergy (ASCIA) (accessed 2026-09-12). Sulfonamide Antibiotic Allergy. Patient guidance. Source / DOI.
- Alcohol and Tobacco Tax and Trade Bureau (TTB) (accessed 2026-09-12). Anatomy of a Wine Label. Sulfite declaration section. Source / DOI.