Neonatal unconjugated hyperbilirubinemia is a frequent reason for treatment during the first days of life. Phototherapy is an effective and non-invasive treatment for reducing bilirubin exposure, but it can also produce biochemical changes in neonates. One important effect is a reduction in serum calcium. Beşer et al. demonstrated significant reductions in total calcium, ionized calcium and parathyroid hormone (PTH) after phototherapy, particularly with longer exposure.¹ An interventional study by Ali et al. also documented a significant decline in serum calcium after 48 hours of phototherapy in neonates with unconjugated hyperbilirubinemia.² These findings indicate that alteration of calcium homeostasis is a relevant biochemical consequence of phototherapy rather than an isolated laboratory finding.

The reported frequency of hypocalcemia after phototherapy, however, varies considerably between studies. Pereira et al. reported hypocalcemia in only 2.5% of term neonates, despite observing a decline in calcium levels in more than half of the study population.³ In contrast, a study conducted in Lahore by Saeed et al. found hypocalcemia in 34.5% of term neonates after 48 hours of phototherapy, with mean serum calcium decreasing from 9.33±1.09 mg/dL to 7.64±1.36 mg/dL.⁴ Javaid et al. subsequently reported hypocalcemia in 33.3% of 150 neonates after 48 hours of phototherapy in Rawalpindi.⁵ This variation suggests that the frequency may depend on population characteristics, gestational maturity, duration of phototherapy and the threshold used to define hypocalcemia.

More recent evidence continues to show inconsistent estimates. Shirolkar and Verneker found a significant post-phototherapy decline in serum calcium after 24 hours of exposure, although clinically important hypocalcemia was less frequent.⁶ In Pakistan, Pervaiz et al. reported hypocalcemia in 23.1% of 134 term neonates receiving 48 hours of phototherapy for indirect hyperbilirubinemia.⁷ The wide variation across recent studies limits the direct application of results from one population to another. Contemporary evidence from Lahore is also limited for a cohort that includes both term and preterm neonates exposed to a standardized 48-hour phototherapy period. The present study therefore hypothesized that phototherapy would produce a measurable fall in serum calcium and that a clinically relevant proportion of treated neonates would develop hypocalcemia. Establishing the local frequency may help determine whether calcium surveillance should receive greater attention during prolonged phototherapy.

The objective of this study was to determine the frequency of hypocalcemia in neonates with hyperbilirubinemia undergoing phototherapy.

This descriptive observational study was conducted in the Neonatology Unit, Department of Paediatrics, Gulab Devi Teaching Hospital, Lahore, 2 January 2025 to 2 July 2025. Ethical approval was taken from the hospital. A total of 130 neonates were enrolled through non-probability consecutive sampling after written informed consent was obtained from a parent or legal guardian. The sample size was calculated at a 95% confidence level, using an expected frequency of hypocalcemia of 17.1% and a 6.5% margin of error. Term and preterm neonates of either sex with unconjugated hyperbilirubinemia requiring phototherapy and completing at least 48 hours of treatment were included. Neonates with birth asphyxia, congenital malformations, sepsis, hypothyroidism, hemolytic anemia, ABO or Rh incompatibility, conjugated hyperbilirubinemia, intrauterine growth restriction, maternal diabetes, pre-existing hypocalcemia, calcium supplementation, jaundice appearing within the first 24 hours of life, or a requirement for exchange transfusion were excluded. Hypocalcemia was defined according to the approved study protocol as a total serum calcium concentration <7.0 mg/dL after 48 hours of phototherapy. Demographic and clinical information, including postnatal age in hours, sex, gestational age in completed weeks, birth weight in grams and baseline serum bilirubin concentration, was recorded on a predesigned proforma. Following enrollment, a peripheral venous blood sample was collected under aseptic conditions immediately before starting phototherapy for measurement of baseline total serum calcium, while the serum bilirubin result used to establish the indication for phototherapy was recorded from the clinical laboratory report. Samples were processed in the Chemical Pathology Laboratory of Gulab Devi Teaching Hospital using its routine fully automated clinical chemistry system under internal laboratory quality-control procedures; the hospital publicly confirms the use of semi- and fully automated chemistry analyzers but does not publish the model used for calcium estimation. Phototherapy was administered using the neonatal phototherapy units available in the Neonatology Unit, which the hospital identifies as part of its neonatal facilities. The neonate was positioned beneath the phototherapy source with maximum practicable skin exposure and appropriate eye protection, while feeding, hydration, temperature and general clinical status were monitored during treatment. Phototherapy was interrupted only for feeding and essential nursing or clinical care. Each neonate underwent two study assessments during the same hospital admission: the first immediately before commencement of phototherapy and the second exactly 48 hours after initiation of phototherapy. At the second assessment, another peripheral venous sample was obtained for total serum calcium using the same laboratory testing system, and the neonate was assessed clinically for manifestations of hypocalcemia. Thus, no separate outpatient research visit was required. This pre-treatment and 48-hour reassessment procedure is consistent with the methodology of the closely related reference study. Data were entered and analyzed using IBM SPSS Statistics version 24.0. Quantitative variables were assessed for distribution and summarized as mean ± standard deviation when approximately normally distributed and as median with interquartile range when non-normally distributed; categorical variables were presented as frequencies and percentages. Pre- and post-phototherapy serum calcium concentrations were compared using the paired-samples t-test, while the change in hypocalcemia status before and after phototherapy was assessed using the exact McNemar test. Quantitative characteristics were compared between neonates with and without hypocalcemia using an independent-samples t-test with unequal-variance correction for normally distributed variables and the Mann–Whitney U test for non-normal variables. Associations of hypocalcemia with age group, sex, gestational status, bilirubin category and birth-weight category were assessed using the Pearson chi-square test, with Fisher’s exact test reserved for cells with inadequate expected frequencies. All tests were two-sided, a P-value <0.05 was considered statistically significant, and relevant estimates were reported with 95% confidence intervals.

The quantitative comparison showed no statistically significant difference between neonates who developed hypocalcemia and those who did not with respect to age, gestational age, birth weight, baseline bilirubin, or pre-phototherapy serum calcium. Neonates who subsequently developed hypocalcemia had a numerically lower mean gestational age, birth weight, and baseline calcium, but none of these differences reached statistical significance (all P>0.05). This indicates that no measured continuous baseline characteristic demonstrated a statistically detectable association with hypocalcemia. (Table 1)

Table 1. Quantitative characteristics according to hypocalcemia status (n=130)

IQR = interquartile range; SD = standard deviation. Mann–Whitney U test was used for age and baseline bilirubin because their distributions were non-normal; Welch's t-test was used for the remaining continuous variables.

Hypocalcemia was numerically more frequent among neonates aged >36 hours than those aged 26–36 hours (38.6% vs 27.6%), among males compared with females (42.4% vs 29.7%), among preterm compared with term neonates (39.1% vs 30.2%), and among those with bilirubin >15 mg/dL compared with 12–15 mg/dL (39.0% vs 31.3%). A decreasing numerical frequency was also observed across increasing birth-weight categories, from 40.7% in neonates weighing <2600 g to 29.4% in those weighing >3000 g. However, none of these associations was statistically significant (all P>0.05), so these numerical differences should not be described as independent risk factors. (Table 2)

Table 2. Association of categorical characteristics with hypocalcemia after 48 hours of phototherapy (n=130)

Percentages are calculated within each subgroup. Pearson chi-square test was used because all expected cell frequencies exceeded 5. A P-value <0.05 was considered statistically significant.

Mean total serum calcium decreased from 9.18 ± 0.50 mg/dL before phototherapy to 7.28 ± 0.75 mg/dL after 48 hours, corresponding to a mean paired reduction of 1.90 mg/dL (95% CI, 1.75–2.05 mg/dL). This reduction was statistically significant (paired t=−25.136, df=129; P<0.001). No neonate met the hypocalcemia criterion before phototherapy, whereas 47 of 130 neonates (36.2%; 95% CI, 28.4%–44.7%) met the criterion after 48 hours. The change in hypocalcemia classification was also statistically significant on exact McNemar testing (P<0.001). These findings demonstrate a substantial post-phototherapy reduction in serum calcium, while the single-arm design does not by itself establish causation. (Table 3)

Table 3. Change in total serum calcium and frequency of hypocalcemia following 48 hours of phototherapy

Hypocalcemia was defined as total serum calcium <7 mg/dL. The 95% confidence interval for prevalence was calculated using the Wilson method. All 130 neonates received 48 hours of phototherapy.

Frequency of hypocalcemia was 36.2% after 48 hours of phototherapy in the present study. This is higher than the 25.6% reported by Bhatti et al. in another Lahore-based study of 125 term neonates.⁸ Sohail et al. in Islamabad observed a significant decline in calcium after phototherapy, although none of their 62 term and late-preterm neonates reached their defined hypocalcemic range.⁹ Naghmana et al. in Rawalpindi reported hypocalcemia in 26.7% of 150 term neonates.¹⁰ Shabbir et al. in Mirpur reported a lower frequency of 15.79% after 48 hours.¹¹ The relatively higher frequency in the present study may reflect inclusion of both term and preterm neonates, differences in calcium cut-offs, baseline characteristics, phototherapy systems and duration of exposure. These comparisons indicate that post-phototherapy hypocalcemia is common, but its measured frequency varies considerably between neonatal populations. Serum calcium decreased significantly from 9.18±0.50 mg/dL before phototherapy to 7.28±0.75 mg/dL after 48 hours, giving a mean reduction of 1.90 mg/dL (P<0.001). Ishfaq and Kafi also documented a significant reduction after phototherapy, with decreased calcium levels occurring in 32% of their participants.¹² In a large prospective Egyptian cohort, Tosson et al. found a significant fall in serum calcium during phototherapy and demonstrated that the reduction increased with exposure duration.¹³ Panneerselvam et al. reported a significant post-phototherapy reduction among 104 term neonates treated with light-emitting diode phototherapy.¹⁴ Jain and Ramesh similarly observed a fall from 9.54±0.92 to 8.26±0.70 mg/dL (P<0.001).¹⁵ The consistency of the direction of change across different settings supports a genuine phototherapy-associated decline in serum calcium. Gestational maturity showed a higher numerical frequency of hypocalcemia among preterm than term neonates in the present study (39.1% versus 30.2%), but the difference was not significant (P=0.323). Poudel et al. found the same direction in Nepal, where hypocalcemia occurred in 34.4% of preterm and 21.3% of term neonates; however, the risk ratio of 1.61 had a confidence interval crossing unity.¹⁶ Jha et al. also reported a higher prevalence among preterm than term neonates, 10.0% versus 6.6%.¹⁷ Reduced calcium reserves and relative immaturity of calcium-regulating mechanisms may make preterm neonates more susceptible, but the present sample did not demonstrate an independent statistical association. Birth weight showed a gradual numerical decline in hypocalcemia from 40.7% among neonates weighing <2600 g to 29.4% among those weighing >3000 g, although the association was not significant (P=0.611). Hussein et al. in Baghdad found a positive correlation between serum calcium and birth weight and identified low birth weight as an independent predictor of post-phototherapy hypocalcemia.¹⁸ The difference from the present findings may be related to variations in birth-weight distribution, gestational maturity, duration of treatment and statistical power. Thus, lower birth weight showed a clinically plausible trend in the present study, but a definite association could not be established. Sex showed hypocalcemia in 42.4% of males and 29.7% of females, but this difference was not statistically significant (P=0.131). Kadavakollu and Gopal likewise found no significant effect of sex on the reduction in calcium following phototherapy (P=0.481).¹⁹ This agreement suggests that biological sex alone is unlikely to be a major determinant of phototherapy-associated hypocalcemia, despite the numerical male predominance observed in the present sample. Baseline bilirubin showed no significant relationship with hypocalcemia. Hypocalcemia occurred in 39.0% of neonates with bilirubin >15 mg/dL compared with 31.3% of those with levels of 12–15 mg/dL (P=0.373). Eddala et al. observed greater calcium disturbance with increasing severity of hyperbilirubinemia and longer phototherapy exposure, suggesting that treatment intensity and exposure may partly explain an apparent relationship between bilirubin and calcium.²⁰ In the present study, the similar baseline bilirubin values among neonates with and without hypocalcemia indicate that bilirubin concentration alone did not reliably identify neonates who subsequently developed hypocalcemia. Postnatal age was also not significantly associated with hypocalcemia, despite a higher numerical frequency after 36 hours of life. This finding is biologically reasonable because neonatal calcium concentration changes naturally during the early postnatal period, while phototherapy introduces an additional exposure-related effect. More importantly, the lack of significance indicates that postnatal age should not be interpreted as a predictor on the basis of the present data. The principal finding remains the biochemical change occurring after phototherapy rather than an age-dependent difference. International frequency estimates further demonstrate substantial variation. Abd El-kader et al. reported hypocalcemia in 65% of 40 Egyptian neonates after 48 hours or more of phototherapy, together with significant changes in ionized calcium.²¹ This value is substantially higher than the 36.2% observed in the present study. Differences in sample size, use of ionized rather than total calcium, gestational composition and laboratory definitions may explain this discrepancy. The international evidence therefore supports the occurrence of calcium disturbance while showing that absolute prevalence cannot be transferred directly between clinical settings. Phototherapy duration was fixed at 48 hours for the primary assessment in the present study and could therefore not be evaluated as an exposure-response predictor. Balan and Biswa reported hypocalcemia in 18% of term and late-preterm neonates and demonstrated increasing frequencies with longer phototherapy, reaching 50% among those exposed for more than 48 hours.²² This observation provides a plausible explanation for the clinically relevant frequency found after the standardized 48-hour exposure in the present study. Taken together, the findings indicate that phototherapy is followed by a significant fall in serum calcium and that a substantial proportion of neonates can cross the hypocalcemia threshold after 48 hours. Limitations of this study include its single-center design, non-probability sampling and absence of an untreated comparison group. Total serum calcium was measured without simultaneous ionized calcium or serum albumin estimation, which limits assessment of physiologically active calcium. All participants were assessed after the same 48-hour exposure; therefore, an exposure-response relationship with phototherapy duration could not be determined. The sample size was also insufficient to establish modest subgroup associations with gestational age, birth weight, sex, age or bilirubin concentration. Future studies should use prospective multicenter designs with larger samples and predefined term and preterm strata. Total calcium, ionized calcium and albumin should be measured at baseline and at standardized intervals such as 24 and 48 hours. Phototherapy irradiance, light source and duration should also be documented precisely. These studies could identify neonates at greatest risk and determine whether selective calcium monitoring is more appropriate than routine testing of every neonate receiving phototherapy.

The conclusion should avoid causal wording because this was an observational study without a control group. Phototherapy was followed by a significant reduction in serum calcium, and hypocalcemia developed in 36.2% of neonates after 48 hours. Routine clinical vigilance and appropriate calcium monitoring may therefore be considered in neonates receiving prolonged phototherapy.

FUNDING

None.

CONFLICT OF INTEREST

None.

AUTHORS’ CONTRIBUTION

All authors contributed equally as per ICMJE.

ACKNOWLEDGMENT

None.