Buaboonnam

Disseminated intravascular coagulation (DIC) is caused by excessive hemostatic system activation. The disease leads to consumptive coagulopathy, microthrombi formation, and severe bleeding. Ultimately, it results in multiorgan dysfunction, manifested in conditions such as trauma, malignancy, and sepsis.[1] DIC is responsible for mortality in such conditions in both child and adult patients.[2,3]
Several scoring systems have demonstrated value in diagnosing DIC. The International Society on Thrombosis and Haemostasis (ISTH) DIC scoring system draws upon prothrombin time, platelet count, fibrinogen, and D-dimer levels, and it has been widely studied in mainly adult patients with DIC.[4] The ISTH DIC system can prognosticate the outcomes of patients in critical condition due to sepsis and non-sepsis etiologies.[5,6] Nevertheless, research on the efficacy of the scoring system for the pediatric population is scarce. Therefore, evaluating ISTH DIC scores in children with DIC may assist physicians in predicting clinical outcomes. This study aimed to evaluate and compare the clinical outcomes of children with overt DIC and non-overt DIC using the ISTH scoring system.

This retrospective study was performed on patients aged 28 days to 15 years who had been diagnosed with DIC during admission at Siriraj Hospital, Mahidol University, Thailand, between January 2005 and December 2014. The clinical parameters of the patients at admission were rated using the ISTH DIC scoring system as follows:[4]
•    Platelet count: > 100 000 cells/mm³ = 0 points; between 50 000 and 100 000 cells/mm³ = 1 point; and < 50 000 cells/mm³ = 2 points
•    Prolonged prothrombin time: < 3 seconds = 0 points; between 3 and 6 seconds = 1 point; and > 6 seconds = 2 points
•    Fibrinogen level: > 100 mg/dL = 0 points, and < 100 mg/dL = 1 point
•    D-dimer level: no increase = 0 points; moderate increase = 2 points; and strong increase = 3 points
The patients were classified into two groups: (1) “overt DIC”, for those with ISTH DIC scores ≥ 5, and (2) “non-overt DIC”, for patients with ISTH DIC scores < 5.
Patients who were previously diagnosed with DIC were treated with blood components within 24 hours before the diagnosis of DIC or had incomplete laboratory parameters of the ISTH DIC scoring system were excluded. Specific organ dysfunctions were classified according to the international pediatric sepsis consensus conference[7] as follows,
Central nervous system dysfunction was defined as a Glasgow coma score ≤ 11 or decreased Glasgow coma score ≥ 3 from baseline.
Respiratory system dysfunction was defined as the ratio of partial pressure of arterial oxygen (PaO2) to the fraction of inspired oxygen (FiO2), PaO2/FiO2 < 300 without evidence of cyanotic heart disease and preexisting pulmonary disease or PaCO2 > 65 torr or > 20 mmHg of baseline or requiring oxygen FiO2 > 0.5 to maintain oxygen saturation ≥ 92% or requiring non-elective invasive or non-invasive mechanical ventilator
Renal dysfunction was defined as the serum creatinine ≥ 2 times for age or increased serum creatinine > 2 times from baseline.
Hepatic dysfunction was defined as total bilirubin ≥ 4 mg/dL or alanine transaminase > 2 times from baseline.
Descriptive statistics were used to detail demographic and clinical characteristic data. Continuous data are presented as medians and ranges, while categorical data are reported as numbers and percentages. Pearson’s chi-squared test was used to compare the proportions of groups with categorical data, and Student’s t-test or the Mann–Whitney U test was used to compare medians for continuous data. Univariable and multivariable predictors of death were evaluated using binary logistic regression analysis (backward method), with results presented as the odds ratio (OR) and 95% confidence interval (CI). A probability (P) value < 0.05 was considered statistically significant. All analyses were performed using PASW Statistics for Windows, version 18.0 (SPSS Inc, Chicago, IL, USA).

Of the 67 992 inpatient cases, 244 patients were diagnosed with DIC, giving a frequency of DIC of 0.35. There were 118 male patients (48.4%) and 126 female patients (51.6%); the age group breakdown of 1 month – 1 year, 1-5 years, 5-10 years, and more than 10 years were 71 (29.1%), 78 (32%), 50 (20.5%) and 45 (18.4%) patients, respectively. Infection was the most common cause of DIC (84.8%), with tissue injury being the second most common cause (7%); other causes are detailed in Table 1. Gram-negative bacterial infection was the most common cause of infection-associated DIC. The other causative organisms are detailed in Table 2. Of the 17 patients with DIC secondary to tissue injury and surgery, open-heart surgery for congenital heart disease (9 patients) was the most common cause. Postorgan transplantation ranked second with 5 patients (liver transplantation, 4 patients; kidney transplantation, 1 patient), followed by intra-abdominal surgery (2 patients) and thermal injury (1 patient). There were 7 patients with malignancies; of these, hematologic malignancies (4 patients) were the most common cause (2 patients with acute lymphoblastic leukemia, 1 patient with acute myeloid leukemia, and 1 patient with non-Hodgkin lymphoma). The other 3 patients had solid tumors: 1 with hepatoblastoma, another with neuroblastoma, and the third with an endodermal sinus tumor. Hemorrhage was the most common manifestation (153 patients; 62.7%), with gastrointestinal hemorrhage being the most common site (41.8%), followed by endotracheal hemorrhage (24.2%) and hematuria (7.8%). Thrombosis was diagnosed in 20 patients (8.1%); venous thrombosis was the most common site (45%), followed by peripheral gangrene (40%).

Table 1. Underlying causes of DIC.

Table 2. Causative organisms of DIC.

The laboratory parameters of the patients with overt and non-overt DIC are compared in Table 3. Of the 244 patients with DIC, 179 (73.3%) were diagnosed with overt DIC, and the remaining 65 (26.7%) had non-overt DIC. The 28-day case fatality rate for overt DIC (76%) was significantly higher than that for non-overt DIC (15.4%; P < 0.001). The median time from the diagnosis of DIC to death was 7.1 days (0–37 days), while the median time from diagnosis of DIC to recovery was 14 days (2–61 days). The correlation between mortality and the ISTH DIC scores is illustrated in Figure 1 (R² = 0.89).

Table 3. Comparison of laboratory parameters of patients with overt and non-overt DIC.

Figure 1. Correlation between the DIC scores and the 28-day case fatality rate.

In terms of admission, 209 patients (85.7%) were admitted to the intensive care unit (ICU), whereas the other 35 patients (14.3%) were not. The rates of ICU admission of the patients with overt and non-overt DIC were not significantly different (P = 0.129). Anticoagulant was prescribed for 9 patients; the most commonly used anticoagulant was unfractionated heparin (4 patients), followed by low molecular weight heparin (3 patients) and warfarin (2 patients). Of the 9 patients requiring anticoagulant therapy, 4 had venous thrombosis, 4 had arterial thrombosis, and one prophylactically received an anticoagulant to prevent clotting after cardiac surgery. Recombinant factor VIIa was prescribed for 9 patients. Neither treatment-associated hemorrhage nor thrombosis was observed. The most typical indication was dengue hemorrhagic fever with severe hemorrhage (6 patients). The other indications were cancer with severe hemorrhage (2 patients) and chronic liver disease requiring a postsurgery bleeding prophylactic (1 patient). The factors associated with the death of DIC patients are presented in Table 4, and the laboratory parameters of the decedents and survivors are compared in Table 5. Univariable and multivariable factors associated with death in DIC are detailed in Table 6.

Table 4. Factors associated with death of DIC patients.

Table 5. Comparison of laboratory parameters of decedents and survivors.

Table 6. Univariable and multivariable factors associated with death.

Studies regarding DIC in children are scarce. Oren et al. reported that the frequency of DIC in the pediatric population was 1.12 hospitalized in a Turkey University Hospital,[8] similar to the present study’s finding of 0.35. However, the frequency in children is lower than in adults (34.4).[9] Additionally, the present investigation found that infections were the most common etiology of DIC, which is consistent with previous studies on adults and children.[8,9] On the other hand, the proportion of children in the current study with infection (approximately 80%) is markedly higher than the corresponding levels previously reported for adults (30%–40%).[10] Furthermore, the rate of malignancy-related DIC seems lower in children than adults, which may indicate differences in the etiologies of the disease in children and adults. Prevalent in tropical regions, dengue hemorrhagic fever can cause thrombocytopenia, plasma leakage, and decreased coagulation factors secondary to hepatic derangement, the combined effects of which lead to DIC.[11] Dengue hemorrhagic fever was the most common cause of viral-associated DIC in the present study; however, this phenomenon might be uncommon in countries where dengue is not endemic. Similarly, our investigation determined that tropical diseases such as disseminated tuberculosis and strongyloidiasis also caused DIC. Therefore, physicians in tropical regions caring for patients with such diseases should be aware of DIC as a peculiar clinical manifestation.
The hemostasis in infants, especially neonates, differs from that in adults. The decreased coagulation factors and natural anticoagulants gradually reach the normal level at approximately six months of age, leading to the counterbalance of hemostasis.[12] The prolonged prothrombin time in such patients might not reflect the proper hemostasis. Therefore, in this cohort, neonates with DIC were excluded from the study.
The clinical severity of hemorrhage and organ failure might be related to the etiologies of DIC in adults. Research on adults showed that multiorgan failure was prevalent in infection-associated DI; in contrast, hemorrhage was common in noninfectious-associated DIC.[13,14] Furthermore, the bleeding tendency in these adult studies appeared to be lower than that of the present pediatric study. The different populations and DIC etiologies of the adult and pediatric investigations may account for the variations in the observed clinical manifestations. The present work identified a thrombosis incidence of 8.1%, comparable with other studies on adult and pediatric populations, and neither arterial nor venous sites predominated.[8,15] Consequently, clinical vigilance of thromboembolic complications is needed in both pediatric and adult patients with DIC. In terms of treatment-associated both hemorrhagic and thrombotic complications, such complications were not observed in this cohort; this may result from the scarcity of patients treated with recombinant factor VIIa, tranexamic acid, and anticoagulant.
Concordant with other studies,[16,17] the 28-day case fatality rate of children with overt DIC or those with organ dysfunction requiring advanced organ support was significantly higher than that for children with non-overt DIC. Additionally, mortality was significantly correlated with the ISTH DIC score and the clinical parameters platelet number, prothrombin time, and D-dimer. Similarly, our multivariable analysis revealed that overt DIC and thrombocytopenia below 50 000/mm³ were associated with death. These results substantiate the role of the ISTH DIC scoring system in predicting the clinical outcomes of DIC in the pediatric population, with platelet number possibly being a pivotal clinical factor in prognosticating the risk of death in DIC. Although other factors (underlying diseases and age at diagnosis) were not significantly associated with mortality, preexisting cancer tended to be correlated with mortality. Therefore, patients with a high ISTH DIC score, especially those with preexisting cancer, should be closely monitored, and treatment interventions for underlying diseases should be delivered promptly.
Compared to the score of 5, the ISTH score of 3 demonstrated a better mortality prediction in sepsis-associated DIC.[18] Furthermore, other DIC scoring systems, namely Texas Children’s Hospital criteria[19] and Japanese Association for Acute Medicine criteria[20] were previously described and demonstrated a good prediction of clinical outcome. However, the former system required sequential evaluation by specialists while the latter required the anti-thrombin level, which was somewhat not performed; these scoring systems might not be applicable in our institute. Taken together, the heterogeneity of results and scoring system substantiated the warrant of further investigation of scoring systems in the pediatric population.
This study had some limitations. First, since this was a retrospective study, there is the possibility of missing or incomplete data. Second, given that the population was drawn from a national tertiary referral hospital, where some tropical diseases appeared to be prevalent, the data may not be generalized to other populations or clinical settings; therefore, the prevalence and causation of DIC may vary from other studies.

Outcomes of Overt and Non-overt Disseminated Intravascular Coagulation Using the ISTH DIC Scoring System in Children: A Single-Center Study