INTRODUCTION

Although rare in pediatric age, venous thromboembolism (VTE) has an increasing incidence and is associated with significant morbidity and mortality ^1,2 . Its etiology in pediatric age is multifactorial and includes hereditary and acquired risk factors (RF), the identification of which is crucial for optimizing the therapeutic approach and prevention strategy for complications and recurrence ^2,3,4,5 .

However, the scarcity of studies makes it difficult to develop action protocols, with most recommendations being extrapolated from adults, with inherent limitations ^1,2 .

The objective of this study is the descriptive clinical-epidemiological analysis of cases of deep vein thrombosis (DVT) of the lower limbs, of patients admitted to the Pediatrics department of a Portuguese hospital, between 2004-2016.

METHODS

This was a retrospective, longitudinal observational study of cases of acute DVT of the lower limbs (LL) in children aged between one month and 17 years and 365 days, admitted to the Pediatrics Department of a Portuguese hospital between January 2004 and October 2016. The cases were searched using the ICD-10 coding for venous thrombosis, and cases of deep vein thrombosis of the lower limbs in this age group were selected. Clinical and epidemiological data were retrospectively extracted from the individual electronic file of each patient, maintaining anonymity. Their clinical evolution up to July 2017 was included. Demographic variables, personal and family history of VTE and hereditary thrombophilia, smoking, usual medication, previous illnesses, history of trauma, recent surgery or immobilization (up to three months before admission), DVT clinical status, body mass index, laboratory and imaging tests, and treatment and evolution to date were analyzed. SPSS v20. was used for data analysis, and frequencies, means and medians were assessed according to the variables under study. The study was approved by the Ethics Committee of the hospital in question.

RESULTS

Eight cases were identified, the main characteristics of which are listed in Table 1. All occurred in adolescents (minimum age 14.25 years; mean age 16.37 +/- 3.99 years), 75% of whom were female.

In all cases, more than one risk factor (RF) was identified and in 87.50%, more than two RFs for VTE, whose distribution by sex is shown in Figure 1 below. Overall, the following stand out: use of oral contraceptives (OC) (100% of adolescents), family history of VTE in a first-degree relative (37.5%), hereditary thrombophilia ” majo” (37.5%), immobilization (25%) and smoking (25%; 25% without record); one case (12.5%) occurred in an adolescent with antiphospholipid syndrome (APLS) and previous episodes of DVT, hypocoagulated with warfarin, but with subtherapeutic INR on admission (INR 1.45); isolated cases of acute infection and obesity were identified.

All adolescents had been using OCs for less than a year, with an average of 5.5 months (IQR 1.63-7.5); all were combined OCs: 66.70% 3rd generation (gestodene) and 33.3% 4th generation (cyproterone acetate).

DVT was unilateral in 100% of cases (75% in the left lower limb). The average time from diagnosis to diagnosis was 2.18 +/- 1.46 days (minimum 12 hours, maximum 5 days), and the most commonly reported symptoms were pain (100%), changes in limb volume (75%) and functional limitation (37.5%). Objectively, the most notable symptoms were edema and pain on palpation of the affected area (87.5%), limited mobility and decreased pulse (37.5%). Limb color change occurred in two cases (25%), and there was one isolated case with heat and another with a positive Hommans sign.

Doppler ultrasound confirmed the diagnosis of proximal DVT (involvement of the iliac, femoral or popliteal segments) in all cases, being considered extensive (ilio-femoral-popliteal) in 75%; the occlusion was total in 25%, partial in 25% and mixed in 12.5%.

Laboratory investigation with complete blood count, coagulation study, renal function, ionogram, glucose, TGO/TGP and C-reactive protein (CRP) was performed in all cases upon admission, occurring after the first dose of low molecular weight heparin (LMWH) in 37.5%. Mild leukocytosis was identified in 12.5%, mild neutrophilia in 50% and elevated CRP in 62.5%. There was a decrease in INR, an increase in aPTT or PT in cases already under anticoagulation and in cases associated with prothrombin (PT) mutation and PS deficiency.

D-dimers were requested in the acute phase in 75% of cases, showing an increase in 83%. Antiphospholipid antibodies were normal in all cases.

In all cases, major hereditary thrombophilias were investigated in the acute phase (Table 2), identifying major thrombophilic defects in 37.5%: n=1 FVL mutation in heterozygosity, n=1 PT G20210A mutation in heterozygosity, and n=1 PS deficiency with mutation in the PROS 1 gene in heterozygosity.

All cases were started on dual anticoagulation with LMWH (1 mg/kg 12/12h subcutaneously) and warfarin (0.2 mg/kg/day orally), with the first treatment being maintained for five days and, until therapeutic INR, for two consecutive days (median nine days (DIQ: 5.5-12)). The median length of hospital stay was 11 days (DIQ 6.25-12.75), and all were discharged on warfarin. The total duration of treatment ranged from 6 to 41 months, and was maintained at the time of the last evaluation in three cases. No complications associated with the treatment were observed in any case. In addition to pharmacological treatment, all patients started wearing elastic support stockings during hospitalization, 37.5% mobilized the limb early and OAC were discontinued.

Regarding acute complications, two cases of PTE were identified in female adolescents (25%) at admission. There were no deaths, hemorrhages, early recurrences, or acute venous insufficiency of the limb. Thoracic CT angiography was performed on admission in 37.5% of cases, confirming PE in one symptomatic case and one asymptomatic case with intra-abdominal extension of the DVT, and excluding it in the adolescent without suggestive clinical signs, but with previous APLS and DVT. In both cases, PE was bilateral, affecting three or more vessels in lobar/segmental branches, and in one case the main pulmonary arteries. The symptoms reported were chest pain, dyspnea, and cough.

At the time of discharge, all cases were referred to the immunohemotherapy and vascular surgery consultation, and 50% were referred to the pediatric consultation. The mean follow-up time in the vascular surgery consultation was 17.75 +/- 3.16 months. At 12 months of age, one case presented symptoms and signs of postthrombotic syndrome (PTS) (edema and collateral vessels in the affected limb) and DVT recurred in the contralateral limb in the adolescent with APLS, with a therapeutic INR (2.4).

DISCUSSION

After the neonatal period, adolescence constitutes the second peak in the age distribution of pediatric VTE, with an estimated incidence of 1.1/10,000 adolescents/year ^1,2,6,7,8 , approximately three times higher than the rest of the pediatric population ⁹ . Although rare, its incidence has increased due to improved care for critically ill children, more frequent use of invasive procedures, improved diagnostic techniques, and greater awareness of this pathology ^{1,4,5,6,7,8,9,10} . This trend appears to be more significant in adolescence, for which a 49% increase in the prevalence of VTE has been described between 2001 and 2007 ¹¹ . In this age group, smoking, OCPs, and pregnancy are additional factors for this phenomenon ³ , and therefore it is more prevalent in females (incidence of 1.49 vs. 0.81/10,000 adolescents/year in males) ^1,2,4 .

Unlike adults, in whom 40% of VTE is idiopathic ^4,6 , in pediatric age it is usually multifactorial, with ≥1 RF being described in 90% ² and only 0-5% of idiopathic cases ⁴ , as in this sample. After central venous catheter, responsible for 50-60% of VTE in pediatric age, chronic diseases are the most prevalent acquired RF (70%), with emphasis on cardiovascular disease, neoplasia, neuromuscular disease, nephrotic syndrome, and chronic inflammatory conditions, such as inflammatory bowel disease and rheumatoid arthritis ^3,5,9,12 . In adolescence, in turn, acute conditions – such as infection/sepsis, trauma/major surgery with prolonged immobilization, dehydration and pregnancy/puerperium – and the use of OC, obesity, sedentary lifestyle and smoking seem to be particularly important ^1,3,4,7,8,11 , as represented in the population studied. Antiphospholipid antibodies (lupus anticoagulant, β2 microglobulin and anticardiolipin), present in isolation in primary APLS or in the context of systemic disease, such as systemic lupus erythematosus, significantly increase the risk of VTE and its recurrence, and should be systematically investigated in all cases ^3,12 .

OCs, which were significantly represented in this sample, increase the risk of VTE by approximately 3-4 times, with an incidence of 1:12,500/women/year without OCs vs. 1:3,500/women/year with OCs ^6,12 . The risk appears to be higher in the first 6-12 months of treatment ⁶ and is 2-4 times higher for 3rd and 4th generation OCs compared to 2nd generation ^{OCs 6,13} . The coexistence of other RFs such as smoking, FVL mutation, obesity and major surgery substantially increase this risk ^7,11 , being present in 50% of adolescents. In this context, practical guidelines for their prescription have been developed. These contraindicate combined OCs in cases of acute or previous VTE, personal/family history of hereditary thrombophilia and/or surgery with prolonged immobilization, with progestin-only OCs and the intrauterine device (IUD) with progestin being a lower-risk option, with only the copper IUD considered safe. In asymptomatic cases with a family history (first degree) of VTE, combined/progestin OCs may be used, although the copper/progestin IUD is preferred. In obese women, contraception with estrogens is not recommended, with progestin-only OCs or IUDs being preferred ¹⁴ . In addition, it is recommended to discontinue OCs at least four weeks before major surgery with prolonged immobilization ² , which did not occur in our sample.

Hereditary thrombophilic defects occur in approximately 10–78% of pediatric VTE ² . Although they increase the risk of VTE and its recurrence, their identification does not affect the acute management of the same ^2,5 , and the overall risk conferred is very low (0.07/100,000 children/year), depending on the defect in question, and does not appear to be significant in the absence of other acquired RF ^2,12 . However, the potential implications for treatment time, antithrombotic prophylaxis in high-risk situations, clarification of the acute situation, and family counseling and healthy lifestyles, seem to justify their systematic investigation in unprovoked cases, particularly in adolescence ^{2,10,11,12,15,16} . The most common hereditary thrombophilic defects are the FVL mutation and the PT G20210A gene mutation, present in 5% and 2% of Caucasians, respectively. PC, PS and ATIII deficiency are rarer and are associated with a higher risk of VTE and recurrence ¹⁵ . Elevated factor VIII and hyperhomocystisemia may be hereditary or acquired and also increase this risk ^2,3,9 . Except for genetic mutations and FVIII, whose increase in the acute phase and post-treatment has prognostic value, the remaining evaluation should be delayed to avoid false positives related to the acute phase of the disease and treatment ^6,12,16 , as occurred in this sample. Hereditary thrombophilic defects were identified in 37.5% of cases, all associated with acquired RF, which is in agreement with the literature ¹⁰ . Although common in the general population, the risk of VTE associated with MTHFR and PAI-1 mutations alone has not been proven, and their routine screening is not recommended ¹⁶ . A history of VTE in first-degree relatives was present in 37.5% of cases and constitutes an independent risk factor for VTE, highlighting the importance of this data in the collection of clinical history ⁶ .

The multifactorial pathophysiology of VTE and the risk factor profile identified in this sample thus highlight the importance of its systematic assessment in routine consultations with adolescents, particularly in the planning of risk situations such as hospitalization and/or major surgery with immobilization, and in candidates for OAC ³ . In this population, an early and risk-adapted approach could have led to a change in attitude and possibly in the risk of VTE in 37.5% of cases, the non-prescription of OAC in obese adolescents, smokers or those undergoing major surgery and/or prolonged immobilization, as well as prolonged antithrombotic prophylaxis in the latter.

Regarding clinical findings, the unilateral presentation of DVT with predominance of the proximal segment of the LLL is in agreement with the literature⁶ , as well as the symptoms and the mean time to diagnosis ¹ . Doppler ultrasound is the recommended diagnostic test, which also allows the assessment of the extent and degree of occlusion and control during follow-up ^2,6 . A complete blood count and coagulation study are recommended before starting treatment to screen for coagulation disorders and avoid false negatives associated with the therapy ² , as observed in the three cases in which the collection occurred under these conditions.

In addition to antiphospholipid antibodies, basic biochemical evaluation is indicated to exclude systemic pathology that may be the basis of VTE and to guide the choice of anticoagulant ² , and was normal in all cases. The elevation of CRP in 62.5% of cases, only one of which had proven infection, may be related to the state of local and systemic inflammation associated with DVT.

Most current recommendations for the treatment of DVT in pediatric patients are extrapolated from adults, with inherent limitations ^2,3,5,6 . The objectives are symptomatic relief, promotion of thrombus resolution and prevention of its extension, embolization, recurrence and complications ^2,6 . The recommended first-line regimen is LMWH (1 mg/kg 12/12h) given its high bioavailability, blood stability and subcutaneous use, with less need for laboratory monitoring ^2,6 . This should be initiated with warfarin (initial dose 0.2 mg/kg orally), maintaining the combination for at least five days and until INR between 2-3 on two consecutive days ^2,6 . In the adolescent population, frequent poor adherence to treatment, concomitant habitual medication, dietary variations and occasional but potentially heavy alcohol consumption can make it difficult to control anticoagulation levels with warfarin. Thus, while the efficacy and safety of new oral therapies (e.g. direct factor Xa inhibitors) have not been proven in pediatric patients, these are aspects that should be given particular attention during regular monitoring ^1,3 .

The duration of treatment depends on the etiology ^2,5,15 . For DVT in the context of reversible RF, 3–6 months of anticoagulation are recommended ^2,3,5,15 . In our sample, two cases underwent more than 12 months, taking into account clinical and epidemiological factors, highlighting the importance of a case-by-case decision based on experience. Given the increased risk of VTE and its recurrence throughout life, APLS is considered a major thrombophilia that requires indefinite anticoagulation ^2,3,5,12,15, as in this sample. When VTE occurs in the context of an irreversible RF, indefinite anticoagulation is recommended, and if it is idiopathic, it should be maintained for 6-12 months ³ . In the context of PE, six months of anticoagulation is recommended ³ . Given the still unclear role of hereditary thrombophilia in the risk of VTE and its recurrence, the duration of anticoagulation in this context is not formally defined and may depend on the defect in question and the concomitant presence of other RF ^10,12 . Thus, given that they were heterozygous, the cases associated with mutations in the PT and FVL genes underwent six months of treatment. For the case associated with PS deficit, given the higher risk of recurrence and its coexistence with persistent smoking and a family history of VTE, it was decided to maintain treatment indefinitely. No complications from treatment were observed in any case, reinforcing the safety of these drugs in pediatric age ³ .

In addition to pharmacological treatment, other general measures are recommended to prevent early recurrence and late complications of DVT, including early mobilization of the affected limb and walking, and the use of elastic support stockings for symptomatic relief ^3,6 . In this sample, early walking and mobilization of the affected limb occurred in only 37.5% of cases, which is an important measure to reinforce. On the other hand, although all cases started wearing elastic support stockings during hospitalization, it was not possible to assess their long-term use in 37.5%, and in one case adherence was sporadic.

Although the prevalence of PTE as a complication of DVT is in agreement with the literature (16.0-31.0%), it may be underestimated since its research was not systematic. Clinical suspicion of PTE in children is difficult because it is often asymptomatic ^3,17,18 ; Furthermore, unlike adults, the use of D-dimers and scores to determine the pre-test probability of PE has not been proven in pediatric patients ^1,2,3,17,18 . Therefore, its diagnosis will continue to require a high index of suspicion and will depend on clinical and imaging factors and the experience of the team ¹⁸ .

PTS and recurrence are the main comorbidities associated with DVT, and are particularly important in pediatric patients due to the greater life expectancy after an event with potential disability and deterioration in quality of life in adulthood ⁵ . Therefore, the identification and modification of RFs for these events, and their early diagnosis and early treatment are crucial in follow-up consultations ¹⁹ .

The recurrence of pediatric VTE is estimated at between 5-10% ^2,5,8, being higher in adolescents (19-22%) ^4,5,7,11 and in idiopathic cases ¹⁵ . Furthermore, it depends on the follow-up time and degree of occlusion ⁴ . In this sample, it occurred in the case of SAFL (12.5%) one year after the index event, which maintained hypocoagulation although at therapeutic levels.

PTS is the most common chronic complication of DVT and is characterized by chronic venous insufficiency of the affected limb ^5,19 . It has an estimated prevalence of 10-70% (weighted mean 26%) ^2,15,19 , being mild-moderate in most cases ^2,19 . The greater number of affected vessels, longer follow-up time after DVT, complete vein occlusion, elevated D-dimer and factor VIII levels at diagnosis and after anticoagulation, and non-complete resolution or extension of the thrombus seem to increase the risk of PTS ¹⁹ . Its pathophysiology presupposes chronic vascular alterations, and the clinical presentation should not be confused with the symptoms of the acute phase; therefore, the diagnosis presupposes its presence in two assessments at least three months apart and (6)12 months after the initial event ²⁰ . Its investigation should be carried out at each follow-up consultation by applying one of the two scales validated and modified for pediatric age – modified Villalta Scale and Manco-Johnson Instrument ²⁰ . In the sample studied, only one case presented any sign/symptom of PTS, reinforcing the need for regular long-term follow-up.

The guidance given to patients at the time of discharge highlights the importance of follow-up in immunohemotherapy consultations for anticoagulation control, vascular surgery consultations for assessment of thrombus resolution and regular monitoring of complications, and Pediatrics for a centralized and holistic approach to each case with advice and intervention in modifiable RFs. Follow-up in Pulmonology consultations of PTE cases is important for monitoring complications such as pulmonary hypertension.