A retrospective study of the clinical characteristics of 9 children with pulmonary embolism associated with Mycoplasma pneumoniae pneumonia

Accurate epidemiological data on pulmonary embolism (PE) in children is scarce in both the United States and abroad. The early incidence estimates of PE stem from autopsy studies and report an incidence of 0.05–4.2% [8‐10]. Studies reported an incidence of 8.6 to 57 per 100,000 in hospitalized children [1, 2], whereas the incidence in all children in the community was estimated to be 0.14 to 0.9 per 100,000 [3, 4]. PE in children has a bimodal age of onset, with two peaks in the neonatal period and adolescence, particularly among those with underlying diseases or risk factors [4]. While rare in children, PE can be life-threatening in severe cases, with a reported mortality rate of 26% [11].

In most cases, Mycoplasma pneumoniae (MP) infections are self-limiting. However, in severe cases, infection can lead to multiple systemic organ complications. Apart from respiratory manifestations, approximately 25% of children with MPP also experience a combination of extrapulmonary symptoms [12]. Narita M has classified extrapulmonary manifestations due to M. pneumoniae infection into three categories [13]: the first is the direct type, in which locally induced cytokines likely play a role; the second is the indirect type, in which immune modulation, such as autoimmunity, is thought to play a role; and the third is the vascular occlusion type, in which vasculitis and/or thrombosis, with or without systemic hypercoagulable states, are believed to play a role.

The specific mechanism of thrombosis caused by MP infection remains unknown, but it may be related to endothelial cell injury, autoimmune response, and an imbalance of coagulation and anticoagulation. The cell membranes of MP contain immune substances such as glycolipids and glycoproteins that share antigens with human tissues and can stimulate the production of autoantibodies. Studies have shown that antiphospholipid antibodies (aPL) and antinuclear antibodies (ANAs) produced by autoimmune reactions caused by MP infection play an important role in thrombosis [14, 15]. Examples of antiphospholipid antibodies include antiphospholipid antibodies (ACA), lupus anticoagulant (LA), and anti-2-glycoprotein antibodies. In Liu’s study of 43 children with MPP and embolism, positive rates for LA were 42.1%, 60% for ACA, 64% for anti-2-glycoprotein antibodies, and 50% for ANAs [16]. In our study, five of nine tested positive for LA, five were positive for ACA, three were positive for anti-β2-glycoprotein antibodies IgM, and six were positive for ANAs. Late follow-up revealed that all antibodies were negative, indicating a potential role for the temporary autoimmune disease brought on by MP infection in the etiology of thrombosis. Moreover, hepatocyte damage caused by MP infection, hypoxia, and immune injury can increase endogenous coagulation factor synthesis [17], while the second synthesis of critical anticoagulation factors such as antithrombin III (AT-III), protein C, and protein S is reduced [18]. Cases of elevated coagulation factors and abnormal anticoagulation proteins have also been reported in MP infection combined with embolism [19‐23]. In the current study, three cases had elevated coagulation factor VIII, promoting the body’s hypercoagulable state; three cases had abnormal protein C or protein S activity, preventing the body from performing anticoagulation normally. However, protein C, protein S, and coagulation factor VIII returned to normal during follow-up, indicating a transient imbalance of the coagulation and anticoagulation system caused by MP infection, leading to embolism.

Children who develop pulmonary embolism (PE) may exhibit fewer clinical symptoms than adults and may not display the classic triad of chest pain, hemoptysis, and dyspnea. Unfortunately, diagnosis in children is often delayed, with some studies indicating an average of 7 days from the onset of symptoms to diagnosis [24]. Our study found that PE typically occurred around 2 weeks into the primary illness, and embolic symptoms only emerged in 6 children after their temperature had stabilized and cough had subsided. Furthermore, the time of diagnosis PE did not always coincide with the acute phase of the disease. Early-stage atypical symptoms of PE may be difficult to distinguish from those of primary pulmonary disease, and young children may not be able to accurately report their symptoms. Underlying conditions, such as pneumonia or heart failure exacerbation, can also mask symptoms of PE. As a result, children with abnormal laboratory indicators should be closely monitored for new symptoms, worsening of existing symptoms, or radiating pain at unusual sites, which may indicate the presence of PE.

D-dimer has been proposed as an independent predictor of embolic risk with high sensitivity [25], and a normal level can exclude the possibility of pulmonary embolism [26]. However, D-dimer has low sensitivity and specificity for the diagnosis of PE in children [27]. Although the combination of the Wells score and D-dimer is commonly used to predict PE in adults, and the absence of both factors can rule out PE [28‐30], this approach has not been found to be effective in children [28, 29]. Fibrinogen (FIB) can cause increased blood viscosity, which typically occurs 6–15 days after Mycoplasma pneumoniae (MP) infection ^[13]. This corresponds to the time to embolism in children with MPP in a national study [31]. In our study, all nine children had varying degrees of elevated D-dimer values, and seven had elevated fibrinogen (FIB), with a median time to embolism of approximately 2 weeks from the diagnosis of the MPP. This is consistent with the hypercoagulable state caused by FIB and the time to peak D-dimer levels in the body. Moreover, platelet (PLT) adhesion and aggregation are major mechanisms of PE, and the activation of PLT by the body’s anti-inflammatory mechanism during MP infection, as well as the elevated PLT count in seven cases in our study, is consistent with the results reported by Li et al [32], which suggest that excessive PLT activation may also be a risk factor for PE. Interleukin-6 (IL-6) and C-reactive protein (CRP) are nonspecific markers of inflammation and tissue damage. Studies have shown that high levels of IL-6 and CRP can promote PE [33, 34]. In our study, IL-6 and CRP levels were elevated in all children, respectively. Although elevated levels of IL-6 and CRP can also be caused by various infectious diseases, our study suggests that PE should also be considered in children with these markers.

All nine children in our study were diagnosed with pulmonary embolism (PE) using computed tomography pulmonary angiography (CTPA), a highly sensitive and specific technique that is currently the most common test for confirming the diagnosis of PE [35]. Because of the small size of the emboli and the atypical clinical symptoms in children, embolic signs have sometimes only been detected upon a second review of imaging findings in previous studies [21, 36]. Therefore, particular attention should be paid to the presence of PE during the imaging review of children with MPP who present with symptoms or abnormal laboratory indicators. The presence of thrombus should be given high priority when evaluating imaging results in such children. In an analysis of 18 children with MPP, solid lung changes (> 2/3 of all lobes) and pleural effusion were identified as independent risk factors for embolism [37]. Notably, all of the children in our study had pleural effusion and pulmonary atelectasis.

There are currently no specific management recommendations for pediatric thromboembolism, and treatment protocols are mainly based on clinical experience and evidence-based research in adult patients. Anticoagulation is the primary therapy for children with hemodynamically stable pulmonary embolism (PE), and fast-acting anticoagulants like unfractionated heparin (UFH) and low molecular weight heparin (LMWH) are recommended as initial treatment [38]. After initial parenteral anticoagulation (UFH, LMWH, etc.), oral anticoagulants such as warfarin should be added. During warfarin therapy, the international normalized ratio (INR) should be monitored, and an INR maintained at 2–3 is recommended. It is generally advised that warfarin be used concurrently with UFH or LMWH for more than 5 days and that the INR be maintained at 2–3 for at least 2 days before discontinuing the medication. However, due to their high dietary impact, difficulty in laboratory monitoring, and teratogenicity, vitamin K antagonists have limited use in children [38]. Direct oral anticoagulants (DOACs) such as dabigatran, argatroban, bivalirudin, rivaroxaban, apixaban, and edoxaban, which have stronger anticoagulant activity and lower bleeding risk, have recently emerged as new options for antithrombotic therapy [39]. Studies have shown that rivaroxaban treatment reduces the risk of recurrence and lessens the burden of thrombosis without increasing bleeding when compared to standard anticoagulants [40]. Clinical trials on rivaroxaban in children (EINSTEIN Jr children I through III) have all been completed. When compared to the conventional anticoagulant warfarin, rivaroxaban has a lower economic burden, is easier to administer, has a faster onset of action, and requires no monitoring [41]. In this study, sequential treatment with rivaroxaban and LMWH was used to treat all nine children with thrombosis. During the follow-up period, there were no side effects such as bleeding or liver or kidney damage, and late follow-up revealed gradual thrombus absorption.

Rapid thrombus dissolution and increased survival are two benefits of thrombolytic treatment. For adult patients with hemodynamic instability who do not have any evident contraindications, European guidelines recommend thrombolytic therapy as the first line of treatment [42]. However, there are no established treatment guidelines for pharmacologic thrombolysis in children, hence it is only used in cases of severe thrombosis that causes hemodynamic instability or acute organ insufficiency. Systemic thrombolysis can be used for high-risk PE in children if the healthcare provider is skilled in thrombolysis and there is evidence that it lowers the risk of death and recurring PE in adults with high-risk PE [43]. Systemic thrombolysis is not recommended for adults with intermediate-risk PE due to the increased risk of major bleeding [44]. In contrast, the use of catheter-directed thrombolysis (CDT) in children with intermediate-risk PE has been shown to have a better outcome [45‐47], reduce the dose of thrombolytic agents, decrease bleeding-related complications, and may be the first choice [48]. In contrast, Urokinase and recombinant tissue-type fibrinolytic plasminogen activator (rt-PA) are the two most commonly used thrombolytic medications at present. The majority of indications, routes of administration, dosages, and durations of thrombolysis in children are still based on the knowledge gained from adult therapy guidelines. In the present study, urokinase thrombolysis was used for two children with mixed cardiac embolism without any thrombolysis contraindications in order to prevent a large PE caused by multiple thrombi and a large thrombus activity. After thrombolysis, their condition considerably improved.

Children with pulmonary embolism (PE) are at an increased risk of requiring surgical embolectomy. Current European guidelines recommend that surgical embolectomy be reserved for hemodynamically unstable patients who have failed thrombolysis or have contraindications to thrombolysis [42]. However, a growing body of literature supports surgical removal of emboli in submassive pulmonary embolism, which has been shown to reduce the mortality rate to 1.1% in high and intermediate-risk cases [49]. Previous reports have shown that children with moderate pulmonary hypertension who underwent surgical embolectomy did not experience complications during postoperative follow-up [12, 13, 33]. For children with acute PE and contraindications to anticoagulation, the placement of an inferior vena cava filter may be considered, but the available literature on this topic in children is limited, and further research is needed for confirmation.