Phototherapy for newborn jaundice is one of the most common treatments in neonatal care, and it is overwhelmingly safe when properly administered. Still, it is not side-effect-free. The recognized short-term effects include temperature instability, increased fluid loss, loose stools, skin changes, and the need for eye protection to prevent retinal damage. Research over the past several decades has also flagged subtler concerns, from riboflavin depletion to a small statistical signal linking phototherapy to certain childhood cancers. None of these risks come close to outweighing the danger of untreated severe jaundice, but parents and clinicians benefit from knowing what to watch for.
How Phototherapy Works and Why It Causes Side Effects
Phototherapy shines blue or blue-green light onto a baby’s skin, converting bilirubin into water-soluble forms the body can excrete without needing the liver to process them. The principal product formed in the blood is a configurational isomer of bilirubin, which has a half-life of roughly 15 hours. A second product, lumirubin, appears at lower concentrations but clears much faster, with a half-life under two hours, making it an important route for bilirubin elimination.1PubMed. Phototherapy for neonatal jaundice: in vivo clearance of bilirubin photoproducts The catch is that the light energy required to drive these conversions does not limit itself to bilirubin. It interacts with skin, blood components, and even certain vitamins, which is where side effects originate.
Temperature Instability
Overheating is one of the most frequently cited side effects. Older fluorescent and halogen phototherapy units generate considerable heat alongside therapeutic light. In one study comparing device types, every newborn treated with conventional fluorescent lights recorded at least one temperature reading at or above 37.5 °C, and more than three-quarters reached 38 °C or higher.2PubMed. Body temperature changes of newborns under fluorescent versus LED phototherapy LED-based devices run much cooler. A trial comparing LED and conventional units found that only about 3% of babies developed hyperthermia under LED lights versus roughly 14% under conventional units.3Journal of University Medical & Dental College. COMPARISON OF LED PHOTO-THERAPY WITH CONVENTIONAL PHOTOTHERAPY IN THE TREATMENT OF HYPERBILIRUBINEMIA IN NEONATES IN TERMS OF SAFETY AND EFFECTIVENESS
The flip side is that LED devices, precisely because they emit less heat, can tip premature or small babies toward hypothermia. A systematic review and meta-analysis confirmed that LED phototherapy carries a higher risk of hypothermia compared to non-LED devices, while producing less hyperthermia and fewer skin rashes.4PubMed. Light-Emitting Diode (LED) Phototherapy versus Non-LED Phototherapy Devices for Hyperbilirubinemia in Neonates: A Systematic Review and Meta-analysis In practice, staff manage temperature by adjusting incubator settings and monitoring the baby’s axillary temperature frequently, but the risk is worth knowing about, especially in very preterm infants.
Fluid Loss and Electrolyte Shifts
Babies under phototherapy lights lose water through their skin at a faster-than-normal rate. Measurements in preterm infants showed a mean increase of about 26% in transepidermal water loss during treatment, with skin folds like the inner elbow and groin losing the most.5PubMed. Transepidermal water loss and skin hydration in preterm infants during phototherapy This invisible water loss, combined with more frequent loose stools (discussed below), means babies can become mildly dehydrated if feeds are not increased.
Electrolytes shift, too. A prospective study tracking sodium, potassium, calcium, blood urea nitrogen, and creatinine found that all of these dropped as phototherapy duration lengthened, with statistically significant declines after 48 hours regardless of the type of phototherapy device used.6Egyptian Pediatric Association Gazette. Impact of phototherapy type and duration on serum electrolytes and blood glucose in neonatal hyperbilirubinemia: a prospective single-center cohort study Low calcium is probably the most clinically relevant of these, since it can cause jitteriness or irritability in the newborn. The practical takeaway for parents is straightforward: more frequent feeding during phototherapy is standard advice and genuinely matters.
Diarrhea and Loose Stools
Green, loose, frequent stools are extremely common during phototherapy and often alarm parents. The color itself is harmless, a result of bilirubin breakdown products passing through the gut. The looseness, though, has a more interesting explanation than you might expect. Early researchers suspected that phototherapy damaged the enzyme that digests lactose, but a study using hydrogen breath tests found no evidence of lactose malabsorption in babies under the lights.7PubMed. Lactose malabsorption is not a cause of diarrhea during phototherapy
The actual culprit appears to be a combination of two factors. First, bile salt concentrations in the stool rise significantly during phototherapy. Babies receiving treatment had stool bile salt levels well above those of controls, and the infants with the highest concentrations were the ones who developed watery stools.8PubMed. Phototherapy-associated diarrhea. The role of bile salts Second, the colon itself temporarily shifts from absorbing water and electrolytes to secreting them, a change that requires both high bilirubin and active phototherapy to occur. Once the lights were stopped and bilirubin fell, normal absorption returned.9PubMed Central. Diarrhoea in jaundiced neonates treated with phototherapy: role of intestinal secretion The diarrhea is self-limiting and does not need treatment beyond keeping up with fluid intake through feeding.
Eye Protection and Retinal Risk
Newborns undergoing phototherapy wear eye patches for good reason. Animal studies have shown that intense blue light can damage photoreceptors in the retina, and the mechanism involves light absorption by rhodopsin, which triggers a chain of events that destroys cells. The neonatal retina is particularly vulnerable because it is still developing.10Journal of Tropical Pediatrics. Neonatal Eye Shielding during Phototherapy: What Protects the Eye Better?
The quality of eye shielding varies widely. Common materials used in many neonatal units, including plain gauze and white fabric without a foam pad, allow a surprising amount of phototherapy light to reach the eyes. During intensive phototherapy at high irradiance, these materials let through enough light to approximate conventional phototherapy levels at the eye surface, which defeats the purpose of shielding. Research on phototherapy hoods has raised similar concerns: blue-light levels visible from within the shadow of a hood may be high enough to warrant additional eye patches, particularly when lamps are placed at the foot end of the incubator.11Acta Paediatrica. Do phototherapy hoods really protect the neonate? Parents sometimes worry when they see the patches shift, and that worry is justified. Nurses check the patches regularly because even brief unshielded exposure to high-intensity light is undesirable.
Skin Changes
Mild skin rashes are among the most frequently reported side effects across clinical trials. They tend to be transient and resolve once phototherapy ends. More striking but far rarer is the so-called “bronze baby” syndrome, in which the baby’s skin takes on a grey-brown discoloration. This occurs almost exclusively in infants whose livers are not clearing bilirubin breakdown products normally, typically because of cholestasis (impaired bile flow). The photoproducts that would normally be excreted accumulate in the skin instead, producing the distinctive coloring.12PubMed. “Bronze baby” syndrome In most reported cases the discoloration fades gradually after phototherapy stops, though one case report linked the syndrome to elevated tissue copper porphyrins in an infant with severe Rh hemolytic disease who ultimately died, underscoring that the syndrome signals serious underlying liver pathology rather than a simple cosmetic issue.13PubMed. The bronze baby syndrome: evidence of increased tissue concentration of copper porphyrins
Even rarer is a purpuric eruption, where nonblanching, well-defined purple lesions appear on exposed skin while shielded areas are spared. One documented case involved a three-day-old infant who developed lesions on the chest, abdomen, arms, and chin during blue-light treatment.14PubMed Central. Phototherapy-induced Purpuric Eruption in a Neonate These eruptions are a curiosity more than a common concern, but they can understandably frighten parents.
Riboflavin Depletion
Blue light does not just convert bilirubin. It also breaks down riboflavin (vitamin B2), which absorbs light at the same peak wavelength. In one of the earliest studies to document this, 16 of 21 infants exposed to phototherapy developed riboflavin deficiency, and every baby who received 49 hours or more of treatment became deficient. Control infants with similar jaundice levels who were not treated with lights did not develop the deficiency, confirming that the light itself, not the bilirubin, was responsible.15PubMed. Light (phototherapy)–induced riboflavin deficiency in the neonate Riboflavin is essential for an enzyme that maintains glutathione in red blood cells, and when riboflavin drops, that protective system weakens, potentially contributing to red blood cell breakdown.16PubMed. Photodegradation of riboflavin in neonates Whether this effect is clinically significant for most term babies receiving short courses of phototherapy remains debated, but it is one reason clinicians prefer to limit treatment duration to what is needed.
Thrombocytopenia
A drop in platelet count has been reported in a meaningful fraction of babies receiving phototherapy. In a study of 100 newborns, 35 developed thrombocytopenia during treatment, with the majority showing mild decreases. Most cases appeared within the first 24 hours and were not associated with clinical bleeding.17PubMed Central. Incidence of thrombocytopenia following phototherapy in hyperbilirubinemic neonates The mechanism is not entirely clear, though light-induced platelet activation and sequestration have been proposed. In most cases, platelet counts recover after phototherapy stops, and the effect does not require intervention unless the baby has additional risk factors for bleeding.
Effects on the Ductus Arteriosus in Very Premature Infants
This side effect is relevant only for extremely premature babies. In the fetal circulation, the ductus arteriosus is a blood vessel that normally closes after birth. In very premature infants, it often stays open, a condition called patent ductus arteriosus (PDA). There is evidence that phototherapy light can relax the smooth muscle in this vessel, delaying its closure. A randomized trial in extremely premature infants found that shielding the chest area from phototherapy light resulted in a smaller ductus diameter and significantly less need for PDA treatment.18PubMed Central. Effect of phototherapy on the ductus arteriosus diameter in extremely premature infants: A randomised controlled trial For full-term or mildly preterm babies, this is not a concern, but it is clinically relevant in the NICU for the smallest and most vulnerable patients.
DNA Damage and Oxidative Stress
The idea that phototherapy light could damage DNA has generated a fair amount of research, especially as intensive high-irradiance protocols have become more common. The evidence here is mixed and depends heavily on the intensity of treatment and whether the baby is premature or full-term.
One study comparing intensive and conventional phototherapy in term infants found that DNA damage markers and a measure of oxidative stress were significantly higher in the intensive group after treatment.19Journal of Neonatal-Perinatal Medicine. Genotoxic effect of phototherapy in term newborn infants with hyperbilirubinemia On the other hand, a study specifically examining blue LED phototherapy in preterm infants found no increase in a key urinary marker of DNA damage regardless of irradiance level or treatment duration.20PubMed. Blue LED phototherapy in preterm infants: effects on an oxidative marker of DNA damage A more recent study added another layer of complexity: blue LED phototherapy was actually associated with lower oxidative stress and DNA damage markers in preterm newborns, though it had no measurable effect in term babies.21PubMed Central. Effect of blue LED phototherapy on oxidative stress and DNA damage in term and preterm infants with hyperbilirubinemia That last finding is surprising and may reflect the fact that bilirubin itself is an oxidative stressor, so reducing bilirubin more effectively could paradoxically lower oxidative markers even as light exposure adds its own stress. The bottom line is that clinicians are aware of this concern and it reinforces the principle of using the lowest effective irradiance for the shortest necessary duration.
Long-Term Cancer Risk
This is probably the side effect that generates the most anxiety when parents hear about it, and it deserves careful framing. Multiple large studies have found a small statistical association between neonatal phototherapy and childhood cancer, but the absolute risk increase is tiny.
A large U.S. study found that cancer occurred in about 33 per 100,000 infants who had received phototherapy versus 21 per 100,000 who had not. After adjusting for confounding factors, the risk increase for overall cancer was about 40%, with somewhat higher signals for myeloid leukemia and kidney tumors specifically. The absolute risk increase amounted to roughly 9 extra cancer cases per 100,000 treated babies, translating to a number-needed-to-harm of over 10,000.22PubMed Central. Neonatal Phototherapy and Infantile Cancer Two recent meta-analyses broadly confirmed these findings. One estimated an overall 24% increased odds of cancer for phototherapy-treated children, with a roughly 40% increase for blood cancers and about 18% for solid tumors.23Frontiers in Pediatrics. Neonatal phototherapy and cancer risk: a systematic review and meta-analysis The other found comparable numbers, with cohort studies showing an odds ratio of about 1.44 for blood cancers.24Pediatric Research. Risk of childhood neoplasms related to neonatal phototherapy- a systematic review and meta-analysis
Relative risk numbers like 24% or 40% sound alarming until you consider the baseline. Childhood cancer is rare to begin with. Moving from roughly 21 to roughly 33 cases per 100,000 is a real difference in epidemiological terms, but for any individual child, the absolute risk remains very small. Researchers also caution that residual confounding is difficult to eliminate: babies who need phototherapy tend to have higher bilirubin, which itself may be a marker for other risk factors. Still, the consistency of the signal across multiple studies and study designs means the possibility is taken seriously, and it is one more reason not to use phototherapy casually when it is not clearly indicated.
Seizures and Neurological Outcomes
A large Danish registry study found a modest association between neonatal phototherapy and childhood seizures, even after adjusting for bilirubin levels. The adjusted hazard ratio was about 1.22, translating to roughly 2.4 extra seizure cases per 1,000 children over ten years. The risk appeared somewhat more pronounced in boys.25PubMed. Childhood Seizures After Phototherapy However, a separate Danish population-based study looking specifically at epilepsy rather than isolated seizures found no increased risk at all, with an adjusted hazard ratio near 1.0.26PubMed Central. Neonatal phototherapy and risk of epilepsy-A Danish population based study The distinction matters: an isolated febrile seizure is different from an ongoing seizure disorder. The data suggest that if phototherapy has any neurological effect, it is subtle and does not translate into a clear increase in diagnosed epilepsy.
Asthma and Allergic Disease
An intriguing but still-debated association exists between neonatal phototherapy (or jaundice itself) and childhood asthma. A large population-based study found that children who had phototherapy or neonatal jaundice had about 30% higher odds of developing asthma before age 12, even after adjusting for confounders.27PubMed. Confirmed association between neonatal phototherapy or neonatal icterus and risk of childhood asthma Separating the effect of phototherapy from the effect of jaundice itself is tricky, since the two nearly always go together. One proposed mechanism involves the immune-modulating effects of bilirubin breakdown products, but this remains speculative. The association is consistent enough to appear in multiple studies, yet not strong enough to change clinical practice.
Home Phototherapy and Whether Side Effects Differ
As home phototherapy has become more common for infants with moderate jaundice, parents naturally ask whether the side-effect profile changes outside the hospital. The available evidence is reassuring. A systematic review and meta-analysis comparing home and hospital phototherapy found no discernible difference in complication rates between the two settings.28PubMed Central. Efficacy of home phototherapy versus inpatient phototherapy for neonatal hyperbilirubinemia: a systematic review and meta-analysis The main practical risk of home treatment is not a different side effect but rather treatment failure: babies may not get enough hours under the light due to feeding schedules, parental uncertainty, or temperature instability that is harder to manage at home. The readmission rate from home phototherapy averages around 3.5%, and no severe adverse events have been specifically attributed to the home setting.29Pediatric Research. Home and in-hospital phototherapy as comparable select routine approach for neonatal hyperbilirubinemia
LED Versus Older Phototherapy Devices
The shift toward LED phototherapy in neonatal units over the past two decades has changed the side-effect landscape. A multicenter randomized trial comparing LED units to compact fluorescent tubes found that side effects were rare in both groups and included hypothermia, hyperthermia, rash, skin darkening, and dehydration.30PubMed. Light emitting diodes versus compact fluorescent tubes for phototherapy in neonatal jaundice: a multi center randomized controlled trial The trade-off is consistent across studies: LEDs substantially reduce the risk of overheating and skin reactions, but they increase the chance of the baby getting cold, especially in premature infants who already struggle with temperature regulation.31PubMed Central. “Light” on Phototherapy—Complications and Strategies for Shortening Its Duration, A Review of the Literature Neither device type has been shown to cause more or fewer long-term complications than the other, largely because follow-up studies of cancer and seizure risk predate the widespread adoption of LEDs and did not separate results by device type.