The dangerous part of teething is the medicine cabinet

Something is always erupting

A baby grows twenty primary teeth by age three, and the first arrive at four to seven months, in the same window the child begins meeting the ordinary run of colds, roseola, and stomach bugs.¹ Lay those two calendars over each other and a fever and a fresh tooth will land in the same week for reasons that have nothing to do with each other.

The overlap is wide. When Macknin's group followed 125 well infants from the four-month visit to their first birthday, they logged 475 eruptions across 19,422 child-days.² Each tooth trailed an eight-day symptom window, from four days before it broke through to three days after. A tooth, on most days, is recently through or about to be. It is the explanation nearest to hand, which is why it collects blame for so much it does not cause. The American Academy of Pediatrics names the most common misattribution outright: "Though many parents expect a fever when baby teeth start to break through, extensive studies show this is a false symptom of teething."³

The signs that actually track eruption

The prospective cohorts agree on a short, mostly local list. Macknin's parents recorded eighteen symptoms daily and matched each against every eruption. Ten showed a real association: more biting, drooling, gum-rubbing, sucking, and ear-rubbing, more irritability and wakefulness, a facial rash where the drool ran, less appetite for solids, and a mild rise in temperature.² The pooled review of sixteen studies ranks the common ones by frequency. Gingival irritation showed up in 86.81 percent of eruptions, irritability in 68.19 percent, and drooling in 55.72 percent, with some sign present around 70.5 percent of the time.⁴

These associations are real and small at once. In Macknin's cohort no teething symptom appeared in more than 35 percent of teething infants, and none appeared more than 20 percent more often than on the days when no tooth was coming.² That second figure is the one that matters at the crib. Drooling tracks teething, but most drooly days have no tooth behind them, so drool predicts almost nothing about whether one is on its way. A sign can be genuinely associated with eruption and still be useless for calling it.

How much a tooth moves the thermometer

Temperature is the one place the cohorts genuinely disagree. The strongest evidence that eruption can raise temperature comes from two studies. Ramos-Jorge's team took daily tympanic and axillary readings on 47 infants over eight months and found a real rise on the day a tooth emerged (P = .004).⁵ Nemezio's meta-analysis pooled the six studies with usable fever data. It found no association overall (odds ratio 1.32, 95% CI 0.88 to 1.96) but a clear one in the subgroup where fever was measured rectally (odds ratio 2.82, 95% CI 1.55 to 5.14).⁶

Both findings survive scrutiny, and both are smaller than they sound. The rise Ramos-Jorge measured topped out at 36.8 C on the eardrum, inside the normal range and nowhere near a fever.⁵ Nemezio's rectal signal rests on a handful of studies, and the overall estimate has a confidence interval that still crosses one.⁶ Macknin, the largest cohort, tied a mild temperature elevation to teething but found no association with any fever over 102 F, and no teething child in the study ran 104 F or fell seriously ill.² The pooled review of sixteen studies reached the same place: eruption "could lead to a rise in temperature, but it was not characterized as fever."⁴

Ramos-Jorge is also the study that ties diarrhea and rash to eruption, which the larger Macknin cohort and the AAP both reject. On the systemic signs the better-powered work wins.⁵²³ The defensible reading is narrow and firm. A tooth can nudge a baby's temperature a fraction of a degree, and a true fever is not something a tooth does.

The remedies the FDA moved against

If teething itself is mild, the mistake that hurts children is in the remedies. Two of the products parents have reached for carry a documented, serious, and avoidable risk, and the federal record on them is not ambiguous.

Start with benzocaine, the numbing agent in Orajel, Anbesol, and Baby Orajel. In its May 2018 letter to manufacturers, the FDA counted 119 cases of methemoglobinemia linked to topical benzocaine between February 2009 and October 2017. Four were fatal, including a four-month-old given an over-the-counter oral benzocaine product for sore teething gums. Of the 119 cases, 116 were serious.⁷ Methemoglobinemia turns hemoglobin into a form that cannot carry oxygen. It can follow a single application, and infants are the most vulnerable because they make less of the enzyme that reverses it.⁷ Against that, the agency reported it was "unaware of any recent studies demonstrating the effectiveness of benzocaine for relieving teething pain."⁷ The risk is measured and the benefit is not. Do not put benzocaine on a teething baby's gums.

The homeopathic teething tablets follow a harder story, because their maker did not go quietly. An FDA inspection that began in September 2016 found the belladonna alkaloid content of Hyland's Baby Teething Tablets "varied widely" from the label, and at times "far exceeded" it. The agency read that as "a fundamental lack of control over the content of toxic chemicals."⁸ The company stopped domestic shipment but declined to pull what was already on shelves, telling the FDA "the public is amply protected."⁸ In April 2017 the agency classified the matter as a Class I recall, its most serious category, and the products were withdrawn to the consumer level.⁸⁹ Its position on the ingredient was flat: "There is no known safe dose or toxic dose of belladonna in children," and infants clear the alkaloids poorly because their liver enzymes and kidneys are still immature.⁹⁸

1. Sep 20, 2016

   FDA inspection begins⁸

   Inspectors find belladonna alkaloid content in Hyland's teething tablets that varies widely from the label.

2. Oct 7, 2016

   Standard Homeopathic halts domestic shipment⁸

   Manufacturing and shipping of the products stop.

3. Jan 2017

   The company declines to recall stock on shelves⁸

   It tells the FDA the public is amply protected and takes no further action.

4. Apr 7, 2017

   FDA requests a Class I recall⁸

   The agency deems the products a serious health hazard, its most serious recall class.

5. Apr 13, 2017

   Hyland's tablets recalled to the consumer level⁹

   All lots of the teething and nighttime tablets are pulled.

6. May 23, 2018

   FDA documents the benzocaine case record⁷

   Its letter to manufacturers tallies 119 methemoglobinemia cases and four deaths and asks firms to stop marketing the gels for teething.

The measures that carry no such record are dull by comparison. The FDA and the AAP converge on the same short list.¹³ Rub or massage the gums with a clean finger, offer a firm rubber teether or a cold washcloth, and past six months a sippy cup of cold water. Chill the teether in the refrigerator, not the freezer. Frozen, it is hard enough to bruise the gums.¹ Skip the liquid-filled toys and the amber necklaces the FDA has tied to strangulation and choking deaths, along with any gel or tablet on the avoid list.¹³ Pain medicine is usually unnecessary. When a baby is genuinely miserable, the AAP's guidance is to ask the pediatrician about acetaminophen, dosed to the child's weight, and to keep it occasional.³

When it isn't the tooth

For a healthy six-month-old, the evidence moves the household's default one way: expect drool, gum-gnawing, and a few cranky days around each tooth, and meet them with cold and pressure rather than a pharmacy. The same evidence draws a bright line around what teething does not explain. A tooth does not raise a real fever, loosen the stools, or make a baby lethargic, and reading those signs as teething is how care gets delayed. Macknin's authors state the rule plainly: before anyone attributes a sign of potentially serious illness to teething, other causes have to be ruled out.²

None of this substitutes for a pediatrician who knows the child. What the research offers is a cleaner prior: the tooth is rarely the emergency, and the two products once sold hardest against it are the ones the FDA had to act on.⁷⁸