In this study, we found that, out of 55 FT infants who developed NT during the study period, about half of them had severe NT at diagnosis and appeared after 72 hours of life in about two-thirds of cases. The most common causes of NT were neonatal sepsis and a postoperative state. Furthermore, severe NT, when compared to mild/moderate NT, was associated with more morbidity (pulmonary or IVH), needed more platelet transfusions, and had increased mortality.
In our study, severe NT was found in 52.73% of total thrombocytopenic cases. This result was higher than other studies. Gupta et al.  found that severe NT accounted for 34.4% of cases. In another study, 20% of cases were classified as severe NT (6). However, Robert et al.  in a large cohort study, included 11 281 NICU admissions of term or preterm infants over 5 years found only severe NT was identified (2.4%). The reason for a higher incidence of NT in our study was probably because the incidence of sepsis in our group was high. This has been shown in other study as well . In contrast to FT infants, in preterm babies, Christensen et al. found that about 73% of extremely low birth weight neonates, at some time during their NICU stay, had at least a one-time platelet count < 150 × 109/L, and this incidence increased up to 85% among neonates with a birth weight ≤ 800 g . Furthermore in our study, most cases of NT in term infants were late onset (after 72hours of life) in about two-thirds of the cases. This was not in agreement with another study in which the majority (84.1%) had early onset NT, but 76% of cases were born preterm .
NT occurs more frequently in association with certain factors, such as sepsis, birth asphyxia, babies born to mothers with pre-eclampsia and low birth weight, and this was seen in our study as well. In our study, the most common cause of NT was neonatal sepsis in about one-third of the cases. Furthermore, the most commonly isolated organisms in septic neonates were gram negative (E. coli and Klebsiella) in 55% of cases. These results agree well with Ree et al. , as they found that severe NT occurred in 20% of septic neonates and the most commonly isolated organisms were gram negative. The pathogenesis of NT in neonatal sepsis is not completely understood. It has been suggested that, in neonatal sepsis, endothelial damage activates reticuloendothelial removal of platelets. NT occurs as, ultimately, the rate of platelet production falls behind platelet consumption . The second most common cause of NT in our study was a postoperative state. Although the definite causes of postoperative NT have not been established in the literature, many factors have been proposed, including post-transfusion dilution, infection-induced, drug-induced, heparin-induced, immune mediated and others .
In this study, most cases (58.18%) were asymptomatic. The most common presentations, occurring mostly with severe NT, were cutaneous bleeding from previous puncture sites and gastrointestinal bleeding. These results agree with a study by Baer et al. . Furthermore, our results also agree with Park et al. , as they found that gastrointestinal hemorrhage in patients with aplastic anemia and severe thrombocytopenia was recorded in 5% of those for whom the lowest platelet count was 20 × 109/L, compared with 1% of those for whom the lowest count was 20 × 109–50 × 109/L.
We found that pulmonary and IVH occurred exclusively with severe NT. This agrees with a study by Setzer et al. , and Bolat et al.  as they found that lower platelet counts correlated with a higher prevalence of IVH. It was not clear from research, until now, whether NT caused the IVH or it occurred after as a result of consumptive mechanisms. In contrast to our results, Baer et al.  found, in patients with severe NT, no relationship between the lowest platelet count recorded and the presence of pulmonary hemorrhage or IVH. They speculated that factors other than NT are prominent in the pathogenesis of those varieties of neonatal bleeding, such as coagulation disorders. Duppre et al.  found that a cellular and humoral coagulation disorder had more of a role in the occurrence of IVH in neonates than thrombocytopenia.
In our study, there were no statistically significant differences between duration of hospital stay and severity of NT, which did not agree with Resch et al. , as they found the duration of NT is positively related to the severity of NT and the number of subsequent platelet transfusions. Furthermore, in this study, half of the neonates with severe NT required mechanical ventilation. This may explain the bad general condition of these patients, and the actual mortality may not only be related to severe NT but also the original disease, such as sepsis, postoperative state or disseminated intravascular coagulopathy.
The outcomes of NT in our study showed that mortality increased to 10.90% with severe NT. In a study by Resch et al. , a mortality rate of 10.8% was significantly associated with signs of bleeding (P < 0.05) and correlated with an increasing number of platelet transfusions (P < 0.05), but not with the severity of NT (P = 0.4). Furthermore, results from studies by Baer et al.  and Resch et al.  found no relationship between the lowest platelet count recorded and the mortality rate; however, a direct relationship was observed between the number of platelet transfusions received and the mortality rate. In our study, two-thirds of cases with severe NT received at least once platelet transfusion. This may be explained by the fact that ill patients receive more platelet transfusions or as adverse effects of platelet transfusions . Therefore, we used the following restricted guidelines for administering platelet transfusions in our unit: (I) platelet count ≤ 100 × 109/L just going to or just having had surgery or having clinical bleeding, (II) platelet count ≤ 50 × 109/L and unstable (mechanical ventilation or vasopressors), and (III) platelet count of 20 × 109/L and stable [6, 21].
In conclusion, severe NT occurred in about half of cases diagnosed with NT in FT infants and appeared after 72 hours of life in about two-thirds of them. The most common causes of NT were neonatal sepsis and a postoperative state. Furthermore, severe NT, when compared to mild/moderate NT associated with signs of bleeding and pulmonary/IVH, required more mechanical ventilation, needed more platelet transfusions, and had increased mortality.