Severe haemophilia A in a neonate presenting as haemopneumothorax after tracheo-oesophageal fistula-oesophageal atresia repair

Abstract

A male infant with oesophageal atresia and distal tracheo-oesophageal fistula (TEF type C) underwent right thoracotomy and transpleural repair of TEF on day 4 of life. He did not have a family history of coagulation disorders. A preoperative finding of prolonged partial thromboplastin time (PTT)>200 s was overlooked, and he went to surgery. There were no concerns with haemostasis prior to and even during the operation. The prolonged PTT was treated with one 10 mL/kg dose of fresh frozen plasma in the immediate postoperative period. On the fourth postoperative day, the infant developed a right haemopneumothorax, requiring fresh frozen plasma and packed cell transfusions. He was subsequently diagnosed with severe haemophilia A due to intron 22 inversion in the factor VIII gene, with factor VIII level <0.01 IU/mL.

Background

A newborn infant with undiagnosed haemophilia A, presenting initially with a complex condition (tracheo-oesophageal fistula (TEF)), underwent major surgery with an uncorrected abnormal preoperative coagulation profile without intraoperative bleeding. This patient had two distinct conditions: type C TEF and severe haemophilia A which presented with delayed postsurgical bleeding after partial correction with fresh frozen plasma (FFP). Through our experience, we report on the atypical presentation of undiagnosed haemophilia A in a neonate with delayed postsurgical bleeding and highlight the importance of timely investigation of an abnormal partial thromboplastin time (PTT) in a neonate.

Case presentation

This male infant was delivered at 35⁺⁴ weeks after prolonged rupture of membranes and suspected chorioamnionitis to a primigravida mother with a history of gestational diabetes requiring insulin. His growth parameters were appropriate with birth weight of 2540 g and he had a non-instrumental vaginal delivery. A TEF was suspected at 2 hours of life after respiratory distress and inability to pass a nasogastric tube. Chest X-ray was compatible with oesophageal atresia and distal TEF (type C TEF). This appeared to be an isolated defect and he did not have other features of VACTERL association. There was no indication that he had bleeding diathesis prior to repair of his TEF as he tolerated vaginal delivery, vitamin K injection, intravenous access attempts and umbilical artery catheterisation with no excessive bruising, haematoma or bleeding. There was no family history of bleeding tendency or coagulopathy. As part of his preoperative workup for TEF repair, the infant’s PTT was checked on day 4 of life from an umbilical artery line running 0.5 units heparin/mL at 1 mL/hour. His PTT was >200 s before and after heparin neutralisation with heparinase (Hepzyme) while his prothrombin time was normal. Given the incongruency between the PTT and the patient’s tolerance of preoperative procedures, another PTT sample was sent just prior to the operation to recheck the abnormal PTT result. Unfortunately, this result came back just prior to the operation time and was not seen nor followed up on before the patient was sent to the operating theatre. He underwent a rigid bronchoscopy and right thoracotomy via a transpleural approach to close the TEF and to repair the oesophageal atresia on day 4 of life. A 10-French chest tube was positioned near the anastomosis of the oesophagus and exteriorised at the right sixth intercostal space at the midaxillary line. There was minimal blood loss and good haemostasis during the procedure. Postoperatively, he received midazolam, morphine and rocuronium infusions for sedation and analgesia. Due to small amounts of blood-tinged aspirates observed in the endotracheal tube 12 hours postoperatively, a PTT was repeated and remained elevated at >200 s post hepzyme. His chest X-ray was clear at this time without an associated need for escalation in ventilator support. He received his first dose of 10 mL/kg of FFP infusion. Six hours post FFP transfusion, his PTT was 61 s. Normal values of PTT on day 5 of life ranges from 25.3 to 59.8 s to 26.9 to 74.1 s for a term infant and a preterm infant (30–36 weeks gestation), respectively.¹

As per surgeon request, he was sedated, medically muscle relaxed and gently ventilated for 3 days postoperatively to preserve the tracheal repair site integrity. In preparation for extubation, he was weaned from sedation and muscle relaxants on postoperative day 4. That night, the patient started to have frequent desaturation episodes with clear endotracheal tube aspirates. A chest X-ray revealed a right pneumothorax (figure 1). Despite handover information that included a history of a prolonged PTT being investigated, a needle thoracentesis was done without prior FFP treatment. This yielded 24 mL of air but no blood. The patient continued to desaturate and repeat chest X-rays showed a residual atypical pneumothorax with increasing fluid collection and a coinciding drop in haemoglobin of 30 g/L (figure 2).

Figure 1.

Small right pneumothorax, right chest drain placed near oesophageal anastomoses during surgery.

Figure 2.

Evolving haemopneumothorax.

A reporting and learning system for patient safety form was filled out by the team detailing the overlooked preoperative PTT result both preoperatively and prior to needle thoracentesis. Disclosure of the events with the parents occurred.

Investigations

Haematological consultation and subsequent investigations for factor deficiencies confirmed severe haemophilia A with factor VIII level <0.01 IU/mL. Genotype testing for haemophilia A showed an intron 22 inversion in the factor VIII gene, the most common haemophilia A gene mutation.

Differential diagnosis

Differential diagnoses of the postsurgical complication in this case included pneumothorax secondary to assisted ventilation with subsequent haemothorax due to coagulopathy,^{2 3} dislodgement of surgical site healing clot due to pressure changes post decompression, bleeding caused by needle thoracentesis and TEF repair anastomoses bleeding or wound dehiscence.

Treatment

Given the previously abnormal PTT results, he was given two 15 mL/kg boluses of FFP, 15 mL/kg of packed red blood cell transfusion and started on Wilate (von Willebrand Factor/Coagulation Factor VIII Complex) 50 units/kg Q12H, and tranexamic acid 10 mg/kg Q8H intravenous. A chest tube was placed to drain the haemopneumothorax. He received daily von Willebrand Factor/Coagulation Factor VIII Complex 20 units/kg while recovering from the episode of haemopneumothorax.

Outcome and follow-up

He was extubated without issues a week later with no further incidents of significant bleeding. Given his diagnosis, he has received appropriate preventative treatment prior to any further invasive investigations or procedures in follow-up for his TEF repair.

Discussion

Haemophilia A affects around 1 in 5000 male births,⁴ while oesophageal atresia with fistula occurs in 1 of 3000–5000 births.⁵ Thirty to fifty per cent of newly diagnosed cases of haemophilia are sporadic; that is, there is no known family history. A Swedish study demonstrated that of 45 cases of sporadic severe haemophilia A, a de novo mutation was found in 17 cases, and the remaining cases from a maternal carrier (mother (23), maternal grandmother (5)).^{6 7} In the absence of a family history of haemophilia, as in our case, it may be challenging to identify severe haemophilia during the newborn period, as there may not be any overt signs of bleeding. Protocols exist for those babies born with positive family history of haemophilia A including fetal sex determination, genetic screening and counselling, and early liaison between obstetric, neonatal and haematology to minimise perinatal and postnatal bleeding risks.⁶ Most undiagnosed neonates with haemophilia present with prolonged or excessive bleeding after a medical procedure such as intramuscular vitamin K administration, blood sampling for metabolic screening, male circumcision or less commonly, intracranial bleeding.^{8 9}

A literature search did not reveal any reports of severe haemophilia associated with TEF or oesophageal atresia. The study by Richards et al noted two cases of postoperative surgical bleeding, after pyloric stenosis and jejunal atresia surgery, as initial presentations of haemophilia in the neonatal period.¹⁰ Interestingly, these are also both gastrointestinal malformations. There were several cases of atypical presentations of haemophilia in the neonatal period reported. One infant presented on day 5 of life with abdominal distension, scrotal ecchymosis and haematocrit of 6.2% due to splenic rupture of undetermined cause. Another neonate developed pseudoaneurysm of the radial artery after arterial blood sampling.^{11 12} Our patient with undiagnosed severe haemophilia presented with right haemopneumothorax 4 days after tracheo-oesophageal repair. Prior to this complication, he had received FFP postoperatively for a PTT>200 s.

Our patient underwent a non-instrumental vaginal delivery. Recent cohort studies of neonates and infants with haemophilia have found intracranial bleeds occurring in 0.6%–3.7% of newborns delivered via this manner.^{10 13} A screening head ultrasound confirmed that he did not suffer from this complication. He also underwent vitamin K administration and venipuncture attempts, reported to be sites of initial bleeding for haemophilia A in 3.6% and 2.4%, respectively, in a surveillance study, without suspicion of a bleeding tendency.¹³ The Swedish study by Ljung et al reported 47/117 (40.2%) of moderate-to-severe haemophiliac newborns exhibited bleeding tendency in the perinatal period, with 24/117 (20.5%) after surgery or venipuncture.¹⁴

One should be wary of overlooking abnormal clotting results in the context of clinical absence of bleeding in a patient who is already receiving high acuity care. In our case, even though our patient had two instances of PTT>200 s, the significance of the markedly prolonged PTT was overlooked in the gamut of tests and intensive care support that the patient needed postoperatively. Additionally, all non-urgent procedures should be preceded by a check for contraindications, including coagulopathies. Our patient should have had a factor assay done earlier to investigate the persistently elevated PTT.⁸ There were 3 days between the postoperative dose of FFP and development of complications; thus, a window of opportunity to investigate and prophylactically manage a coagulopathy and potentially avert the haemopneumothorax was not realised. In this case, an urgent, instead of routine consult and discussion should have occurred with the consulting haematology service.

In our patient’s case, he may have tolerated the TEF repair without complications as adequate primary haemostasis relies on small amounts of thrombin generation through the tissue factor pathway, platelet aggregation and function.¹⁵ The delay in bleeding was likely due to partial postoperative treatment with FFP which contains 0.70–0.87 IU/mL of factor VIII.¹⁶ Receiving 10 mL/kg of FFP postoperatively would have meant he received approximately 20 IU of FVIII (8 IU/kg) raising his FVIII level to 16% at a peak of 1 hour but decreasing over the next 8–16 hours. This FFP was enough to slow down and delay major bleeding but inadequate to completely prevent it especially in the face of additional invasive procedures.

A review of the patient safety forms filed about this case has resulted in a preoperative checklist that is completed and reviewed by the team prior to every patient going to the operating theatre.

In review, a case of a coexistence of two rare conditions, TEF with oesophageal atresia and haemophilia, is described. Additionally, it illustrates that confirmed abnormal PTT requires timely investigation and management even in the absence of bleeding during a major surgery. All non-urgent invasive procedures should be considered high risk and be delayed, if possible, until the severity of coagulopathy can be ascertained as prophylactic management and haematology input is important.

Patient’s perspective.

As first time parents we did not even know what occurred with our baby could happen. I was put on modified bed rest at 22 weeks due to a large fibroid. My baby’s nuchal scan had been normal, and while blood tests did show a risk of 1 in 92 for trisomy 18, we did not go for amniocentesis or fetal DNA testing. Our 18 week ultrasound scan and subsequent scans did not show any signs that indicated anything was wrong with our son. Our son was born early at 35+4 weeks and when the doctors explained what they believed was wrong with our newborn—that he had a defect that prevented him from swallowing properly, we were in shock and complete disarray. The care team in our initial hospital and the subsequent tertiary surgical NICU were caring and they were always ready and willing to speak to us and give us their time to answer questions.

A few days after TEF/EA surgery, our baby developed the bleeding around the lung. The doctor came to explain that in fact, our baby had blood tests showing suspected clotting issues preoperatively. Plasma was given after surgery but the doctor explained that they should have caught and managed the condition sooner. To us, the way the surgery went was amazing as there were no bleeding episodes during the procedure. We did not find out that our son’s Factor VIII levels were so low until quite a few days after. I have since been tested and the results show that I am not a carrier and we can 100% say haemophilia A is not in my family and was not inherited. The biggest thing we take away from this is that even with no genetic conditions in our family history, anything is possible. We have a special boy who lives with two conditions and not just one. This is something we constantly remind the specialists who manage him. We not only worry about food getting lodged in his throat, chronic reflux and chronic respiratory issues, but we also worry about him falling and having severe hemorrhages and all that comes with haemophilia.

Learning points.

• All surgical patients should have coagulation results reviewed and confirmed by the medical team prior to the operating theatre with system checks in place to ensure this cannot be overlooked.

• A persistently abnormal coagulation profile should be urgently investigated even in the setting of a unremarkable medical history including tolerance of major surgery without bleeding.

• A neonate with undiagnosed severe haemophilia may go through major surgery without excessive bleeding; however, marked bleeding in the postoperative period or with subsequent procedures may develop.

• Two rare conditions, such as a tracheosophageal fistula with oesophageal atresia and haemophila, can coexist in the same patient.