Severe Hypernatremic Dehydration and Lower Limb Gangrene in an Infant Exposed to Lamotrigine,Aripiprazole,and Sertraline in Breast Milk

Introduction

Hypernatremic dehydration is rare in breastfed neonates. Its incidence has been estimated at 0.07% to 4.1% of term neonates.^1–4 In severe cases, hypernatremic dehydration can lead to life-threatening complications. Here, we report a case of a breastfed neonate with severe hypernatremic dehydration whose mother was treated with lamotrigine, aripiprazole, and sertraline.

Case Presentation

In 2013, a 12-day-old exclusively breastfed male neonate presented with severe weight loss and hypernatremic dehydration because of inadequate milk intake. He was born at term by spontaneous vaginal delivery. He was the couple's second child. Delivery was uneventful. Birth weight was 3.896 g and Apgar scores were 9-9-9 at 1, 5, and 10 minutes. The mother and infant were in good condition and left the hospital 44 hours after delivery. Discharge weight was 3.645 g.

The infant's mother was treated for bipolar disorder with lamotrigine 250 mg orally once daily, aripiprazole 15 mg orally once daily, and sertraline 100 mg orally once daily. She had already been taking lamotrigine and aripiprazole before pregnancy at these doses; they were continued during pregnancy and postpartum without dose modification. Sertraline was added at the beginning of pregnancy and continued in the postpartum period. Other medications were levothyroxine 50 μg once daily, a prenatal multivitamin, and folic acid. Naproxen, docusate, and the combination of acetaminophen and codeine were taken only on the first postpartum days.

Based on information recorded in the medical chart, 3 days before hospitalization, the infant slept for about 8 to 12 consecutive hours. The following day, he was irritable and sleepy. A visit for weight evaluation was scheduled at 12 days of life; the weight was recorded as 2.710 g (30% weight loss since birth) and the infant was referred to the emergency department at a tertiary care hospital.

On initial evaluation in the emergency department, he was pale, with marbled skin, dry mucous membranes, decreased skin turgor, and bluish feet (the right foot eventually became darker with blackened toes) with prolonged capillary refill. He was moaning and weakly responsive. His heart rate was 157/min, blood pressure 66/39 mm Hg, respiratory rate 100/min, and O₂ saturation 100% on room air. He had no fever. Blood glucose was 7.9 mmol/L (142 mg/dL). A jugular access was put into place and he received fluid resuscitation with isotonic fluid, as well as empiric therapy with ampicillin and gentamicin.

He was admitted to intensive care unit (ICU) for hypovolemic shock and suspected right lower limb arterial thrombosis. Initial laboratories showed severe hypernatremia (Na 199 mEq/L), normal potassium (4.3 mEq/L), mixed acidosis (capillary blood gas: pH 6.95, pCO₂ 74.0 mm Hg, bicarbonate 15.9 mEq/L), acute renal failure (blood urea nitrogen 67.0 mmol/L [188 mg/dL] and serum creatinine 359 μmol/L [4.06 mg/dL]), and disseminated intravascular coagulation (hemoglobin 17.2 g/dL, platelets 136 × 10³/μL, white blood cell count 15.9 × 10³/μL; international normalized ratio (INR) 1.73, prothrombin time 19.3 seconds and activated partial thromboplastin time (APTT) 36.3 seconds, fibrinogen < 60 mg/dL). Alanine aminotransferase was slightly increased at 70 U/L; aspartate aminotransferase and gamma glutamyl transferase were normal (28 and 32 U/L), as well as total bilirubin (16 μmol/L [0.94 mg/dL]) and alkaline phosphatase (145 U/L). Serum glucose was slightly increased at 13.2 mmol/L (238 mg/dL). Serum lactate was 4.20 mmol/L (37.84 mg/dL). Plasma lamotrigine concentration was measured and found to be 13.0 μmol/L (3.33 μg/mL).

Cerebral CT scan showed petechial lesions and a mild decrease in cortico-subcortical differentiation, without sinus thrombosis. Abdominal ultrasound was normal. On Doppler ultrasound, no thrombi were visible in the right lower limb, but it was impossible to view blood vessels below the right knee. Bone scan showed hypoperfusion of the right lower limb below the distal half of the tibia, with severe hypoperfusion of the right foot phalanges.

In ICU, volemic repletion continued. Antibiotics were adjusted to ampicillin and cefotaxime. Further therapy given in ICU included heparin, fresh frozen plasma, cryoprecipitate, platelet and packed red blood cell transfusions, as well as vitamin K and topical nitroglycerin (1% ointment).

The infant was hospitalized for a total of 52 days. Renal failure was corrected within 6 days and hypernatremia after 7 days. Other laboratory anomalies also progressively normalized. Blood and urine cultures were negative. The most significant long-term complication was gangrene of the right lower limb, which did not respond to medical therapy and required amputation of all five toes and surgical debridement of the metatarsals on day 19 of hospitalization. Necrosis was attributed to arterial microthrombi caused by disseminated intravascular coagulation after severe dehydration. Thrombotic work-up identified no underlying signs of coagulopathy.

Other complications during hospitalization included rhabdomyolysis and possible brief partial seizures with secondary generalization at 38 days of life. Cerebral MRI was normal and EEG showed diffuse slowing without epileptic activity. No treatment was given and there were no relapses.

Development was normal other than mild gross motor delay at 5 months, according to a rehabilitation specialist, and gross motor development was below the fifth percentile at 13 months on evaluation by a physiotherapist. The infant did not receive breast milk after hospital admission and discharge.

Discussion

Hypernatremic dehydration in breastfed neonates (serum sodium ≥150 mEq/L) is usually recognized between 6.5 and 10 days of life.^1,2,4,5 Our patient was diagnosed at 12 days and had been progressively lethargic since day 9 of life.

The incidence of this complication varies according to the methodology used in published studies. One group estimated it at 0.7/1,000 breastfed neonates. ¹ The incidences of breastfeeding-associated hypernatremic dehydration among newborns otherwise range from 1.9% to 4.1%.^2–4 In these studies, serum sodium at presentation was 150 to 194 mEq/L and weight loss since birth was 3% to 33%.^1–4 Serum sodium as high as 214 mEq/L has been reported by others. ⁵ Our patient, with a serum sodium of 199 mEq/L and 30% weight loss at presentation, had thus developed severe hypernatremic dehydration.

Severe complications of hypernatremic dehydration among exclusively breastfed neonates are described in Table 1. Peripheral gangrene has also been described in 14 infants with hypernatremic dehydration because of other causes (gastroenteritis and febrile illness).^6–9

Hypernatremic dehydration in breastfed infants is typically reported in previously healthy term neonates, after normal pregnancy and delivery, and can occur with mothers who have successfully breastfed in the past.^1,2,5,10,11 As reported in other studies, our patient's hypernatremic dehydration was because of unsuccessful breastfeeding.^1,2,4 Numerous references report parents being unaware of the progressive malnutrition and dehydration because signs are nonspecific (quiet infants who sleep well and seem to have a normal urine output).^1,2,5,10

Our patient's mother was treated with lamotrigine, aripiprazole, and sertraline. Previous case reports of hypernatremic dehydration leading to gangrene in neonates have not reported maternal medication use.

Aripiprazole is the least known of these treatments. Based on data from four women treated with 10 to 18 mg/day, it is estimated that an exclusively breastfed neonate would take <8% of weight-adjusted maternal dosage, which is usually acceptable.^12,13 Insufficient milk production is described in three women treated with aripiprazole: one woman failed to establish lactation, one woman exclusively breastfed her infant for 6 weeks, then needed milk supplementation (her serum prolactin was lower than expected), and one woman noticed decreased milk production within 24 hours of introduction of aripiprazole with a significant increase 48 hours after treatment cessation (causality is unclear because aripiprazole has a prolonged half-life of at least 75 hours).^12–14 Galactorrhea has also been reported in nonlactating patients treated with aripiprazole. ¹² Impact of aripiprazole on prolactin and milk production might change from one patient to another and be difficult to predict because of its unique mechanism of action related to partial agonist activity at D2 receptors. ¹⁴

Transfer of lamotrigine in breast milk has been more studied. In one case series of 29 breastfed infants whose mothers took a mean dose of 304 mg/day, mean level in infants was 7 μmol/L (1.79 μg/mL), with a range from nondetectable to 25.7 μmol/L (6.59 μg/mL). ¹⁵ Others reported plasma concentrations from 1.2 to 15.2 μmol/L (0.31–3.90 μg/mL) among exposed infants. ¹² Our patient had a plasma lamotrigine concentration of 13.0 μmol/L (3.33 μg/mL), which is in the therapeutic range, but compares with what has been reported in breastfed infants without side effects. One incidence of severe apnea and cyanosis has been reported in an infant whose mother was taking a high dose (850 mg/day with a corresponding plasma concentration of 58 μmol/L [14.87 μg/mL] in the mother and 19 μmol/L [4.87 μg/mL] in the infant). ¹⁶ Causality has been debated in this case. The rate of adverse events during breastfeeding is low with this medication. ¹⁷

Sertraline use is well described during breastfeeding, usually leading to low or undedectable plasma concentration in infants.^12,18 Side effects were only reported in 2 infants on 279 exposed during lactation in a recent review. ¹⁸ Even though codeine has been incriminated as a high-risk medication if taken by an ultrarapid metabolizer at CYP2D6, a large body of information provides reassurance for short-term use such as our patient's mother. ¹²

Our patient's mother's polytherapy is of interest. Reports about the use of medication during breastfeeding usually give data on individual medications, yet we know that polytherapy can lead to additive side effects.

Our case adds to the knowledge about use of these medications in lactating mothers, even if causal inference is difficult to assess.

The probability that the mother's medication might have caused the adverse clinical event was assessed using the Naranjo scale. ¹⁹ We calculated a total score of 3 points on a maximum of 13, which categorizes that the adverse reaction was caused by the medications in the category of “possible” (1–4 points). Points were allocated for the following items of the scale: the adverse event appeared after the suspected drug was administered (2 points), the adverse reaction improved when the drug was discontinued (1 point), there are alternative causes (other than the drug) that could on their own have caused the reaction (−1 point), the adverse event was confirmed by an objective evidence (1 point).

Conclusion

Severe complications of hypernatremic dehydration in exclusively breastfed neonates are rare but can be devastating. Awareness of these complications should prompt close follow-up of the infant with poor weight gain. The role of maternal medication as a risk factor for hypernatremic dehydration among exclusively breastfed infants needs to be further explored. Adding this case to the pooled experience of maternal medication use while breastfeeding can improve counseling of these mothers.