Does breathing steam help colds and sinus congestion?

The idea

The theory is that warm, moist air helps thick mucus drain and that heat could damage cold viruses, as it does in the laboratory¹. In a test tube, rhinovirus needed about an hour at 43°C to be inactivated³.

What the trials show

The Cochrane review covered six small trials with 387 people, testing heated air from a machine (the RhinoTherm) rather than a bowl. It could not say whether steam relieves cold symptoms: one analysis favored it and another did not. Viral shedding did not differ, and some people had nasal irritation¹. In two double-blind trials of 66 and 68 patients, steam did no better than room-temperature air^2,3. An earlier trial reported about half the symptom score after 20 minutes of 43°C air compared with 30°C air, but later trials did not repeat it⁴. We found no randomized trial of a room humidifier for colds; a review of indoor humidity found complex effects on comfort and on virus survival¹².

The scald risk

Burns centers in the Netherlands, Wales, England and Australia all report children scalded by steam inhalation. Many cases involve a spilled bowl or a vapor device, and children needed skin grafts more often than adults^5,6,7,8. One group of authors points to children's limited motor skills, curiosity and poor awareness of danger⁶.

Safer ways to get comfort

Saline nasal rinsing possibly relieves symptoms, though trials were small, so it is a reasonable option; see nasal irrigation¹⁰. Menthol, camphor and eucalyptus vapors give a cooling sensation of easier airflow without changing nasal resistance, which is why they feel helpful¹¹; see eucalyptus. Sitting in a closed bathroom with the shower running briefly is the alternative suggested by the authors of a Brisbane burns study⁷. For burn first aid, see minor burns and wounds.

Croup

Steam is often used for a child with croup (a barking cough with noisy breathing). In a randomized trial of 140 children with moderate to severe croup, humidity did no better than room air⁹.

Do heated vapor trials show a benefit?

The 2017 Cochrane review included six randomized trials from five publications with 387 participants. For symptom relief, a fixed-effect analysis favored heated humidified air (odds ratio 0.30; 95% CI 0.16 to 0.56; 2 studies, 149 participants), but the random-effects analysis did not (OR 0.22; 95% CI 0.03 to 1.95), so the effect was called uncertain. The number reporting resolution of symptoms was not significantly higher (OR 0.58; 95% CI 0.28 to 1.18; 124 participants), and viral shedding did not differ (OR 0.47; 95% CI 0.04 to 5.19). Nasal resistance worsened in one US study and improved in an earlier Israeli one. The reviewers rated the evidence low quality and concluded that the data show neither benefit nor harm for heated humidified air delivered by the RhinoTherm device¹.

Macknin and colleagues randomized 66 patients to two 20-minute treatments of saturated air at 40 to 42°C (32 patients) or ambient air at 20 to 24°C (34 patients). The placebo group improved more on some symptom scores on isolated days (placebo versus steam: 40% versus 25% on day 3 and 100% versus 67% on day 7) and in nasal resistance on day 7 (11% versus -6%), and the authors found no benefit². Forstall and colleagues first showed that 43°C for at least an hour inactivated rhinovirus in vitro, then randomized 68 volunteers to a single 60-minute treatment that raised the intranasal temperature to 43°C (32 patients) or ambient air (36 patients). Daily symptom scores over seven days did not differ (P = .59 to .83)³.

The positive trial

Tyrrell and colleagues ran a double-blind trial in 87 general-practice patients and 84 volunteers, comparing fully humidified air at 43°C with air at 30°C. Patients treated for 20 minutes at 43°C had roughly half the later symptom score of those at 30°C, and volunteers treated for 30 minutes on three occasions had an 18% reduction in symptoms. But a regimen of 20 minutes at onset, then 10 minutes daily, showed no difference⁴. The later Cleveland trials, with a similar approach, did not reproduce the benefit^2,3.

How often does steam burn people?

In the Netherlands, a prospective database found about three people a year admitted to burn centers for steam inhalation burns, mostly children, who needed skin grafts more often than adults⁵. In Swansea, 16 children (average age 7.4 years) attended a burns centre over five years; burns averaged 3.1% of body surface, and 17 of 21 surveyed GPs still recommended steam, including 8 of 19 who recommended it for children under five⁶. In Brisbane, 27 children were scalded in six years: 44% were under three, with hand burns from vapor devices in the youngest and larger spill scalds in children aged five to fourteen; most burns were deep dermal partial thickness, two children needed grafts, and no child suffered an airway scald⁷. A London burns centre treated 19 patients in two years, six of whom needed surgery, and national data identified 201 cases over 2018 and 2019⁸. These are hospital case series, so true rates are unknown, and probably higher⁸.

Croup, saline, menthol and humidity

Scolnik and colleagues randomized 140 children aged 3 months to 10 years with moderate to severe croup to blow-by humidity, 40% humidity or 100% humidity with particles sized for the larynx. Changes in the croup score at 30 and 60 minutes did not differ between groups⁹. The saline irrigation Cochrane review found five trials (544 children, 205 adults) compared with routine care rather than placebo; most results showed no difference, one larger pediatric trial showed small improvements in nasal secretion and obstruction scores (mean differences -0.31 and -0.33 on a four-point scale) and less decongestant use, and the authors judged the trials small and at high risk of bias¹⁰. Burrow and colleagues found that five minutes of camphor, eucalyptus or menthol vapor left nasal resistance unchanged, although most subjects felt cold air and improved airflow¹¹.

Bottom line

Confidence is moderate to low (3 of 5). Trials of steam show no consistent benefit for colds, and the single positive trial was not repeated^1,2,3,4. The harm, scalds from hot-water bowls and vapor devices, is real and falls mostly on children^5,6,7,8. Comfort measures with a better safety record are the better bet^7,10.