Abstract
Purpura fulminans (PF) is a rare and fatal complication of septic shock or diffuse intravascular coagulation (DIC) resulting in skin and soft tissue necrosis. PF can be caused by congenital or acquired protein C (PC) or protein S (PS) deficiency. The most common cause of PF in a neonate is sepsis. In our extremely low birth weight preterm case, due to PF that started in the right-hand fingers, examination was made and protein S deficiency was detected as well as MTHFR (A1298C) and Factor V Leiden (R506Q) homozygous mutations. While being unresponsive to fresh frozen plasma (FFP) and unfractionated heparin (UFH) therapy, we want to highlight the curative treatment with hyperbaric oxygen (HBOT), which has not previously been used in extremely low birth weight preterm infants for this purpose.
Introduction
Neonatal purpura fulminans (PF) is a rare but high mortality thrombophilia that causes hemorrhagic skin necrosis and widespread intravascular coagulation. It may occur due to congenital (protein C, protein S deficiency) or acquired causes (such as infections, DIC, antiphospholipid antibodies, galactosemia, severe congenital heart failure, warfarin treatment). The most common cause of neonatal PF is sepsis [1–3]. The recommended treatment is fresh frozen plasma (FFP) and anticoagulant therapy with the appropriate antibiotics. Hyperbaric oxygen therapy is used effectively in newborns, especially in treating necrotic tissue lesions [4, 5]. Here, we present an extremely low birth weight (ELBW) preterm infant, followed in the Cerrahpasa Medical Faculty neonatal intensive care unit (NICU), with protein S deficiency presented as PF that was unresponsive to adequate FFP and unfractionated heparin and treated with hyperbaric oxygen.
Case
She was born with weight 940 grams with cesarean section due to premature labor at 30 + 2 gestational weeks to a family of non-consanguineous 44 year-old healthy father and 37 year-old mother with MTHFR (heterozygous) and Factor V Leiden (heterozygous) mutation. Her Apgar scores were 5 and 7 in the first and fifth minutes respectively. On her first postnatal day, surfactant was given for the diagnosis of respiratory distress syndrome as well as ampicillin plus gentamicin antibiotics therapy as empiric treatment. She was supported by nasal CPAP. Umbilical vein catheter was inserted and total parenteral nutrition was started. Due to CRP elevation (18.5 mg/L) and thrombocytopenia (49.000/mm3) on postnatal 3rd day with the diagnosis of early neonatal sepsis, antibiotic treatment was changed to sulperazone and amikacin. On postnatal 4th day, paleness and bruising that started in the right-hand distal phalanges was noticed and progressed rapidly (Fig. 1a). Blood pressure values were within normal range, right radial and ulnar artery pulses were palpable, and flow was observed naturally by doppler ultrasonography (USG). Her blood examination showed that platelets (PLT): 33.000/mm3 and CRP: 9.27 mg/L. There was no bleeding on cranial USG, and a small atrial septal defect was detected on echocardiogram. Elevation and heating of the hand as well as local nitroglycerin treatments were administered considering venous thrombosis. Upon proximal spreading of the lesions, the case was consulted with Pediatric Hematology and Oncology Department. Her coagulation test results were; prothrombin time: 14 seconds, APTT: 54.5 seconds, INR: 1.1, Fibrinogen: 179 mg/L, D-dimer: 3.93 mg/L. She was diagnosed with PF and treated with heparin loading dose which then followed by infusion (15μ/kg/hour) and twice a day FFP (10 ml/kg) treatment. Blood cultures were found negative.

Photos of the patient’s right hand; a-Pre-treatment, b-Second day of the HBO treatment, c-End of the treatment.
Although maximum doses (heparin 25/kg/hour, FFP 4×10 ml/kg/day) were reached on the 3rd day of the treatment, amputation was started to be considered as the ecchymotic-cyanotic appearance continued to progress on her fingers. After consulting to Underwater and Hyperbaric Medicine Department of the Istanbul University, hyperbaric oxygen therapy (HBOT) was applied at 1.4 atmospheres absolute (ATA), two hours each and as four treatment cycles. On the 2nd day of the HBOT, a moderate improvement was observed in two fingers (Fig. 1b). Except for the second and third distal interphalangeal joints of the right hand, the patient fully recovered (Fig. 1c), and was discharged with only low molecular weight heparin (Clexane) once a day subcutaneous treatment on postnatal 75th day.
In the examinations performed at the time of diagnosis of the mother, protein S was low at 39.8% (Normal: 60– 130), protein C was 119.4% (N: 70– 140), and antithrombin-3 was 107.9% (79– 112). In the postnatal 6th month of our patient, only protein S was 27.9% low (Normal: 60– 130), MTHFR (A1298C) and Factor V Leiden (R506Q) homozygous, PAI-1 heterozygous mutations were detected. At the postnatal 6th month; her neurological development was compatible with 4 months. She was breastfeeding. Her head circumference, height and weight were observed in 10, 10, and 10– 25 percentile respectively. She didn’t have premature retinopathy or intraventricular bleeding during follow-up.
Neonatal PF was first described as a hereditary disease in 1962 by Van der Horts. Hereditary PF is associated with complex heterozygous or homozygous deficiency of protein C and S, also can be seen as a result of other congenital thrombophilia [1, 2]. Recurrent pregnancy losses are common in babies born to mothers with gene anomalies [6, 7]. The fact that our patient’s mother had 3 pregnancy losses, due to a reason that could not be determined previously, suggests a genetic pathology that predisposes thrombophilia. The genetic test of our patient showed MTHFR (A1298C) and factor V Leiden (R506Q) homozygous and PAI-1 heterozygous mutation. Culture negative, early-onset sepsis and underlying protein S deficiency were thought to be the cause of thrombophilia. In this sense, our case is similar to the case of protein C deficiency secondary to early neonatal sepsis presented by Tina et al. [2].
Sepsis is the most common and well-defined cause in PF. While commonly group B streptococcal infections were seen in the newborn period, cases of PF related to gram negative bacterias were also observed [8–10]. Our case was diagnosed as a culture-negative clinical sepsis due to CRP elevation and thrombocytopenia development one day before the PF presentation. In the literature, it is shown that PF develops in the clinical sepsis due to the endothelial damage causing the disruption of synthesis and/or increased consumption of coagulation factors [1, 2]. From the neonatal period until postnatal sixth month, coagulation cascades are not effective enough and there is low plasma PC-PS activity. In addition, there are many factors effecting serum PS-PC levels, therefore, repetitive tests as well as genetic examination are required for definitive diagnosis [11–13]. In our patient, MTHFR (A1298C) and factor V Leiden (R506Q) were found homozygous as well as low PS level at the postnatal sixth month. Both in the acute period and at 6 months, the PS levels of the mother and the baby were found to be low, supporting the inheritance. Children with severe hereditary PC-PS deficiency have a lifetime risk of developing PF and therefore require long-term anti-thrombotic therapy with PC replacement and/or anticoagulants [14, 15]. Since this etiology could not be excluded in our patient, she was discharged with low molecular weight heparin treatment.
Hyperbaric oxygen therapy (HBOT) indications
If PF occurs secondary to sepsis, appropriate antibiotics should be started immediately and sepsis management should be done. For the congenital or acquired thrombophilia secondary to sepsis, FFP is administered as well as anticoagulant therapy or activated protein C concentrate (APC). Mortality and morbidity were found to be high even in early diagnosed and treatment-initiated PF cases. In most of them, unwanted interventions such as amputation needed to be done [2, 16]. In our case, sepsis was recognized early and antibiotic therapy was given in addition to unfractionated heparin and FFP treatments. Although the blood circulation and clinical status of our patient did not deteriorate, the necrosis in the right hand fingers progressed. Amputation was considered but not required with the help of HBOT as the last chance, which stopped the progression of the necrotic area in the right-hand fingers and even healed.
HBOT is a technique in which the patient is exposed to 100% oxygen for a certain time and at a certain pressure in special rooms. Oxygen pressure should be at least 1.4 ATA or higher. The indications for this treatment as can be seen in Table 1; are carbon monoxide poisoning, chronic wounds, ischemia, and soft tissue lesions due to infections in children and adults [4, 18]. Antimicrobial effects have also been described, especially in cases of PF caused by an underlying infection. Complications of HBOT are barotraumas (e.g. pneumothorax), visual complications due to hyperoxia, and rarely hearing disorders. In the literature, it is shown that side effects other than barotrauma are generally caused by reactive oxygen species (ROS) and this effect is reversible [19, 20]. It is known that the high oxygen concentration for a long time in premature infants can cause retinopathy of prematurity. However, studies on HBOT found to be safe in applications that do not exceed 3 ATA and 2 hours, but side effects due to barotrauma and oxygen toxicity were observed on 4 ATA [19–21]. In our case, the Istanbul University Underwater and Hyperbaric Medicine Department, that has a large number experience with neonates and heparinized patients, carried out a 4-day treatment without exceeding the safe limits. As a result, the necrotic area regressed and complications (such as retinopathy and intraventricular hemorrhage) did not occur.
HBOT has been used successfully in the treatment of PF that rapidly progresses to necrosis and has high mortality and morbidity rates. As we presented in our case, HBOT has been shown to be an effective treatment method in extremely low birth weight premature infant when applied carefully, for the appropriate time and dosage. Our suggestion should be supported with additional studies for the final recommendation.
