VTE is a common complication in patients with cancer. PE is a form of VTE that is common and a major cause of morbidity and mortality in cancer patients. Patients with HM constitute a particular subset of cancer patients with significant differences. Thus far, little is known about the differences between HM and tumor patients in terms of the clinical characteristics at the time of the development of PE. PE can be classified into different subgroups based on the temporal pattern of presentation (acute, subacute, or chronic), the presence or absence of hemodynamic instability (massive PE, submassive PE, or low-risk PE), the anatomic location (saddle, lobar, segmental, or subsegmental), and the presence or absence of symptoms. The signs and symptoms of PE are often nonspecific, making the diagnosis particularly challenging. The clinical presentation ranges from mild dyspnea or chest pain to sustained hypotension or shock. Sometimes, PE may even be asymptomatic and consequently diagnosed only with imaging procedures performed for other purposes. According to the results of 904 tumor patients with PE in the MD Anderson Cancer Center emergency department (ED) [12], symptoms were observed in 20% of patients, and shortness of breath (17.3%) was the main symptom. VTE was discovered in 20.9% of the patients with PE as a concurrent incidental finding and was associated with poor overall survival. In a study reported by the People's Liberation Army (PLA) 307 Hospital with 52 Chinese tumor patients included [13], symptoms of PE were observed in 88% of the patients. A total of 53.1% of the patients complained of dyspnea without obvious causes. A total of 34.4% of the patients had DVT. The main tumor types in the ED were gastrointestinal system tumors (29.4%) and respiratory system tumors (46.2%) in Chinese patients, which might have caused a lower proportion of patients with dyspnea in the ED study and a higher proportion of patients with dyspnea in the PLA study. Despite the large number of cases included in both studies, only a small proportion were HM patients, 9.9% and 23.1% respectively, and data on HM patients were lacking. In this study, symptoms were observed in all HM patients and 94.9% of tumor patients. In addition, 80% of the HM patients and 72.2% of the tumor patients presented with dyspnea, indicating that the majority of Chinese cancer patients with PE had clinical manifestations and that dyspnea might be the first sign in both Chinese HM and cancer patients. This is different from the results reported in Western countries, which may be caused by different constitutions of cancer types and different physical qualities of patients.
In addition, 27.9% of the patients in the ED had central PE and 36.4% of the patients had multiple levels of PE at the same time. A unilateral embolus location was recorded in 71.8% of patients. At PLA 307 Hospital, 54.2% of the patients had central-type PE, and the number of patients with multiple levels of PE was not specified. However, in the present study, the incidences of central PE were significantly lower and the incidences of multiple levels of PE were much higher in HM and tumor patients than those reported by the ED. Significant differences were not found between HM and tumor patients with respect to the location of PE. Further multi-center analysis with a large sample size are needed. But, it is worth noting that patients with acute leukemia had a higher incidence of peripheral PE, whereas patients with NHL had a higher incidence of central PE. Most of the cases involved unilateral PE. These differences may be caused by the nature of the disease itself.
VTE is likely to be a symptom of occult malignancy, and PE is often the first manifestation in patients with tumors[14]. Natasha Kekre summarized the incidence of VTE in patients with HMs [15], reporting that the overall risk of VTE in patients with ALL and APL remained high at close to 10%. Patients with AML had a VTE incidence of 5–8%, and patients with aggressive lymphoma had a VTE incidence of 4.2%. In a single-center study of over 1000 patients [16], patients with high-grade lymphoma had almost twice the incidence of VTE compared with that in patients with low-grade lymphoma (10.6% vs. 5.8%). Besides, the latest study reported by Veli Bakalov[17] analyzed the risk factors for VTE in 4236 hospitalized patients with hematological malignancy in the US. The largest number of PE patients with hematological malignancy to date, that is, a total of 944 patients, was included. According to its results, the rate of VTE was highest in patients with AML (6.6%), followed by patients with ALL (6.1%) and NHL (6.0%), and the VTE rate was lowest in patients with MM (3.5%) and chronic lymphocytic leukemia (CLL) (3.3%). The top three incidences of PE among the hematological malignancies occurred in CLL (28.7%), NHL (26.8%) and ALL (25.3%). However, in this study, more than half of the HM patients had malignancies derived from B-cells. NHL, especially DLBCL, had the highest incidence among the reported cases rather than AL or CLL. None of the cases were CLL or chronic myelogenous leukemia (CML) in this study. The incidences of VTE were similar between HM and tumor patients. Combining all of the above data, PE was more common in NHL than in other types of HM. But the types of HM most prone to PE might be related to the population in different areas. What’s more, it showed that a high incidence of VTE does not necessarily mean a high incidence of PE in HM patients.
Fifty percent of tumor patients had PE occurring in the 3 months before or after the tumor diagnosis[13]. The highest incidence of VTE occurred in the first 6 months from diagnosis in patients with ALL, AML and APL and within the first 12 months in patients with CLL and lymphoma [15].In this study, the onset time of PE in tumor patients had two peaks. The majority of tumor patients developed PE within 2 weeks or 3 months after the diagnosis of the cancer, which was basically consistent with the literature. Patients with gynecological tumors had a tendency to develop PE within the first month after diagnosis, which was much earlier than in patients with other types of tumors. And, most HM patients developed PE within 50 days of the diagnosis of HM. This showed that PE in HM patients occurred much earlier than in patients with most other types of cancer. Patients with HM or gynecological tumors need to be alert to the risk of early PE when developing acute dyspnea.
MM has a high rate of VTE associated with immunomodulatory therapies, with this rate ranging from 1–6%[18]. The use of thalidomide and lenalidomide has been associated with a marked increase in the risk of VTE[18]. The study reported that the highest incidence of VTE occurred in the first 16 months from diagnosis in patients with MM and in the first 3 months in patients treated with immunomodulatory drugs such as thalidomide or lenalidomide [15]. In this study, patients with MM had a PE incidence of 8.3%. Consistent with the VTE incidence in the above-reported investigations, all of the patients in the present study experienced PE in the first 3 months.
Tumor metastasis aggravates a high blood-coagulation state, which is a major cause of the increasing rate of PE in patients with advanced cancer [15]. The strongest risk factors for VTE other than cancer are infectious complications, including sepsis, invasive candidiasis, pneumonia and IV line infections [17, 19]. In a study reported by Wang H[13], factors such as age, sex, smoking history and hypertension were not significantly different between tumor and nontumor patients with PE, but factors such as coronary heart disease, hyperlipidemia, chronic obstructive pulmonary disease and diabetes were significantly different between the two groups. In this study, the medical history of all patients with PE had no obvious similarities. However, 64% of HM patients had pneumonia when PE occurred and 40.5% of tumor patients had surgical history within 30 days. Among tumors patients who had a surgical history within 30 days, 46.9% developed pneumonia. In addition, 68% of HM patients in this study were male; moreover, all of the male patients with PE had a history of smoking, and 76.5% of the male patients had pneumonia. The results indicated that developing pneumonia and a history of smoking in male patients were potential risk factors for PE occurrence in HM patients, especially in DLBCL patients. For patients with tumors, a surgical history within 30 days complicated with pneumonia might increase the risk of PE occurrence.
Furthermore, chemotherapy was an important trigger for hospitalization secondary to VTE [20]. The odds ratios for the development of VTE were significantly higher in tumor patients receiving chemotherapy than in chemotherapy-naïve cancer patients [21]. The possible mechanism for this increased risk might be direct damage to endothelial cells caused by chemotherapy and surgery, induction of the coagulation pathway or activation of the coagulation system in vivo by the release of tissue factors. This study showed that ongoing chemotherapy in HM patients was associated with a greater risk of PE than tumor patients. In addition to chemotherapy, CVCs also constitute a very important clinical risk factor for VTE [22]. According to a study reported by the MD, CVC-PE events were responsible for 9.4% and 4.7% of cases in the ALL and AML groups, respectively [22]. In our center, a PICC line was usually placed once a patient was diagnosed with HM. A total of 36% of the HM patients had PICCs and most of them developed PE within 50 days after the PICC line was placed. However, it was difficult to identify the effects of malignancies and PICCs on PE occurrence.
In addition, several risk factors for VTE in cancer patients, such as thrombocytosis, leukocytosis, low hemoglobin levels, and elevated D-dimer levels, have been reported [23–26]. According to PE guidelines, a D-dimer level below 0.5 µg/mL is a PE exclusion criterion. The D-dimer level was found by Bai CM to be elevated in 90.9% of PE patients [27]. Similarly, the D-dimer level was elevated in 84% of HM cases and 97.5% of tumor patients in the present study. However, the D-dimer values in HM patients were much lower than those in tumor patients. Although the hemoglobin levels in HM patients were much lower than those in tumor patients, this finding might be related to the HM. To date, there are no clear data supporting the association of the abovementioned factors with PE in patients with HMs. Multicenter studies with a larger number of PE cases are needed for further analysis.
Treatment of PE in HM patients is more complex than that for PE in the setting of other kinds of cancer. Anticoagulation in patients with hematologic malignancy is challenging due to severe thrombocytopenia and a high risk of bleeding. Some studies have indicated that withholding anticoagulation when platelets drop below 50×109/L could avoid bleeding but is associated with higher VTE recurrence [28–31]. Limited data are available to guide decisions on the management of PE prophylaxis, and treatment in HM patients. In this study, 80.0% of HM patients received anticoagulant or thrombolytic therapy in the hospital. No treatment-related mortality or major bleeding was observed. The 30-days PE-related mortality rate was similar in HM and tumor patients, and no significant difference was found.
Our investigation was limited because it was a retrospective and descriptive single-center study. The number of PE patients included was limited. In addition, because CTPA was not regularly performed in our center, some PE patients without clinical symptoms might have been missed. In addition, the long-term bleeding rates and incidence of recurrent VTE were not captured by this study because patients received outpatient medical treatment after discharge from the hospital leading to loss to follow-up in some of them. Therefore, a long-term clinical observational study should be performed in PE patients with HMs.