What could cause bright red blood vomits in a neonate?

11 Flares Twitter 8 Facebook 2 Google+ 1 LinkedIn 0 Email -- Filament.io 11 Flares ×

A 6 day old infant male vomits a large amount of bright red blood at home and is taken to a rural emergency department. The child looks good, but the amount of blood on the baby’s blanket brought in by his mother is very impressive.

The E.D. physician estimates it to be 30 to 50 cc’s.

He was born to a 25 year old G7P1 Blood type AB+, rubella immune, syphilis negative, hepatitis B negative mother at 37 weeks gestation via spontaneous vaginal delivery without any risk factors for sepsis. Apgar scores were 8 and 9. He was circumcised on day 2 of life without any complications or excessive bleeding.

Mother had a history of substance abuse but a drug analysis at birth was negative. Mother had visited the child’s pediatrician once and he was assessed as healthy at the time. He is currently breast feeding. He is passing stools and urinating well.

In the rural E.D., a guaiac of the red material on the blanket that the baby vomited was positive. The baby passed a dark stool in the E.D. that was also guaiac positive.

A CBC was drawn and the infant was transferred to a children’s hospital by ambulance.

Exam findings on arrival

VS T37.3 (rectal), P128, R35, BP 75/50. He is alert and active in no distress. Color is pink with minimal jaundice. Anterior fontanelle flat and soft. TM’s within normal limits for age. No blood in the nares. Oral mucosa pink and moist. No bleeding or mucosal injuries noted. Conjunctiva pink (no pallor). Neck supple.

Heart regular without murmurs. Lungs clear. Abdomen soft, flat bowel sounds active, without hepatosplenomegaly. The umbilical cord is drying without bleeding. No hernias. Normal testes. Healing circumcision. Peripheral pulses good. Normal visible perfusion. Capillary refill time 1 second. No petechiae or bruising visible. Stool is guaiac positive.

An abdominal series is ordered. In the meantime, his mother is observed feeding him with a bottle and he is noted to feed very well. The CBC done at the rural E.D. is normal. His hemoglobin is 17.3, hematocrit 49.6. A repeat of his hemogram shows Hgb 17.2, Hct 50.3. Coagulation studies are normal. A urine toxicology screen on the infant is negative.

An Apt test done on the hematemesis residue on the blanket is negative for fetal hemoglobin indicating that the blood on the blanket did not come from the infant. The abdominal series is cancelled. However, some “positive” neonatal radiographs are shown below.

Neoabdo1

This abdominal radiograph shows radiographic evidence of necrotizing enterocolitis (NEC). The presence of air in the intestinal wall (intramural air) is known as pneumatosis intestinalis (arrows). This is seen as a double density layering of the intestinal wall sometimes called “railroad tracks” as opposed to the single layer density of a normal bowel wall.

In neonates, pneumatosis intestinalis is highly indicative of NEC. NEC presents in the neonatal period with signs and symptoms that include guaiac positive stool, poor feeding, vomiting, hematemesis, abdominal distention, abdominal discoloration, abdominal tenderness, bowel obstruction, or other findings suggestive of an acute abdomen. Premature infants are at highest risk. It is unlikely to occur in healthy term infants.

This is generally not a condition that presents as an outpatient to an emergency department. It is most often diagnosed by neonatologists in intensive care level low birth weight infants. Infants with NEC should be managed in a neonatal special care unit. They are at risk of sepsis, bowel necrosis and/or perforation and may need the expertise of a pediatric surgeon.

Despite this, pediatricians and neonatologists may occasionally refer outpatient newborns to an emergency department to investigate the possibility of NEC in a newborn who was previously discharged. Thus, it is important to have seen some of the positive radiographic findings to properly assess the radiographs of such infants.

neoabdo2

 This radiograph shows generalized bowel dilation suggestive of an obstruction. The bowel walls appear to be smooth (loss of haustrations). Additionally, this radiograph shows another radiographic sign of NEC. Note the faint air luncencies over the liver. This indicates the presence of intraportal air which is highly indicative of NEC.

neoabdo3

In addition to bowel dilation suggestive of an obstruction, this radiograph shows a more obvious case of pneumatosis intestinalis. A large segment of bowel in the patient’s left lower region (cigar shaped) shows obvious intramural air along its length.

Although the physicians involved in our infant’s care were concerned about the possibility of NEC, it is unlikely that NEC would present with bright red hematemesis as it did in this instance. Benign causes of hematemesis are largely due to the ingestion of blood. An intra-oral or nasal injury resulting in some swallowed blood will often be followed by hematemesis. Mothers occasional have breast injuries due to epidermal erosions from the intense suckling of the newborn infant. These can hemorrhage. The infant can swallow this blood while breast feeding.

In this patient’s case, his mother’s nipples were noted to be bleeding. This was felt to be the source of the bleeding since the Apt test was negative for fetal hemoglobin. The Apt test is not ordered frequently by emergency physicians, but it may be very useful in instances such as this to rule out the infant as the source of the bleeding.

Although the rural E.D. physician estimated the volume of hematemesis to be 30 to 50 cc’s, the actual volume of hematemesis was probably only 10 cc’s. This was determined by spilling some colored liquid onto a blanket to reproduce the area of hematemesis noted on the infant’s blanket. This accounts for all of the infant’s findings except for the guaiac positive stool. However, a newborn’s meconium stool is normally guaiac positive. If the Apt test was done at the rural E.D., it is likely that the infant would not have required transfer to another hospital.

Pulmonary hemorrhage may occasionally present with hemorrhaging in the mouth that may resemble hematemesis. Pulmonary hemorrhage results in respiratory distress and insufficiency usually requiring oxygen and/or positive pressure ventilation. Mallory-Weiss syndrome has been reported in infants. Forceful or prolonged vomiting results in a laceration of the esophagus diagnostically confirmed on endoscopy. Esophageal varices resulting from portal hypertension may be caused by thrombosis, hepatic fibrosis, or congenital malformations of the portal circulation. Thrombosis may be a complication of exchange transfusion or umbilical vein catheterization. Esophageal varices may spontaneously bleed, resulting in large amounts of blood loss. Gastric and duodenal ulcers may occasionally occur in newborns. In this age group, peptic ulcer disease usually presents with hematemesis.

References

Vanderhoof JA, Zach TL, Adrian TE. Gastrointestinal Disease. In: Avery GB, Fletcher MA, MacDonald MG (eds). Neonatology Pathophysiology and Management of the Newborn, fourth edition. Philadelphia, J.B. Lippincott Company, 1994, pp. 605-629.

Gryboski J. The Esophagus. In: Gryboski J. Gastrointestinal Problems in the Infant. Philadelphia, W.B. Saunders Company, 1975, pp. 48-117.

Fanaroff AA, Filston HC, Izant RJ. Selected Disorders of the Gastrointestinal Tract. In: Klaus MH, Fanaroff AA (eds). Care of the High-Risk Neonate. Philadelphia, W.B. Saunders, 1993, pp. 176-188.

Print Friendly
11 Flares Twitter 8 Facebook 2 Google+ 1 LinkedIn 0 Email -- Filament.io 11 Flares ×

About 

Prof Loren Yamamoto MD MPH MBA. Professor of pediatrics at the University of Hawaii and a practising pediatric emergency doctor in Honolulu. | Contact | View Loren's DFTB posts