Radiological Findings on Chest Computed Tomography in Patients With the Primary Diagnosed Chronic Lymphoproliferative Diseases

Introduction The presence of concomitant respiratory pathology complicates the process of treatment and recovery of patients with chronic lymphoproliferative diseases (CLDs). Therefore, the diagnosis of lung injury is an important step in the management of such patients. Objectives The aim of this study was to determine the prevalence, nature, extent, and location of changes diagnosed by high-resolution chest computed tomography (CT) in patients with CLDs at the initial examination. Methods Medical records of inpatients who were hospitalized in 2018-2019 to the City Hematology Center of the Public Non-Profit Enterprise “City Clinical Hospital #4” of Dnipro City Council with a confirmed clinical diagnosis of CLDs were included in the retrospective study. The results of initial high-resolution chest CT were studied and analyzed. Results Out of 1,004 hospitalized patients with confirmed CLDs, 119 patients were primarily diagnosed. Among them, 81 patients underwent chest CT examination (68.1%) before the beginning of specific therapy. The average age was 65 (56;68) years, 46 (56.8%) were men. 23 (28.4%) patients were diagnosed with chronic lymphocytic leukemia, 28 (34.6%) patients with multiple myeloma, 24 (29.6%) patients with lymphoma, and six patients (7.4%) had other CLDs. It was found that both central and peripheral lymphadenopathy had about a third of the studied cohort of patients (33.3 and 29.6%, respectively), and these symptoms dominated in patients with chronic lymphocytic leukemia (43.5 and 50%, respectively), lymphoma (50 and 52.2%, respectively), and other CLDs (45.8 and 16.7%, respectively), in contrast to patients with multiple myeloma (7.1 and 0%, respectively). Lesions of the lung parenchyma were found in 45.7% of the studied cohort and were met more often in patients with multiple myeloma (67.9%). However, when comparing the prevalence of their categories, no statistically significant differences were found. Predictable bone-destructive changes were statistically significantly more common in patients with multiple myeloma than in other groups of patients with CLDs (P=0.0003). Conclusions Signs of pulmonary diseases during initial chest CT were found almost in half of the patients with CLDs. It potentially may affect the frequency of treatment complications in such patients. Lymphadenopathy was the most common finding on chest CT, especially in patients with chronic lymphocytic leukemia and lymphoma. And enlarged intrathoracic lymph nodes possibly could lead to pulmonary functions disorders. Among the types of lung parenchyma lesions, pneumofibrosis and foci of consolidation in the lower lung lobes were the most often diagnosed. Chest CT is informative at the stage of the initial examination of patients with CLDs not only for clinical diagnosis but also for the diagnosis of respiratory comorbidities and prediction of the disease outcome and treatment complications.


Introduction
Chronic lymphoproliferative diseases (CLDs) are pathological conditions of heterogeneous nature, the 1 2 factors of development of which are considered to be both environmental factors and genetic pathology [1]. The most common CLD is chronic lymphocytic leukemia (CLL) with a global incidence of up to five per 100,000 population and a higher prevalence among men [2]. Multiple myeloma (MM) ranks second among oncohematological pathologies in prevalence (about four cases per 100,000 population), which increases with age [3]. Various pathomorphological variants of lymphomas are also common among CLDs. For example, non-Hodgkin's lymphoma occurs in 19.7 cases per 100,000 population in the United States, and its prevalence increases with age (from 9.3 per 100,000 under 65 years and up to 91.5 per 100,000 population among people over 65 years) and almost twice more common among men (23.9 per 100,000) than among women (16.4 per 100,000 population) [4].
Management of patients with CLDs is a quite difficult task because in addition to issues of effectiveness, complications, and side effects that occur with different types of treatment are extremely important for the prognosis. A common side effect of chemotherapy and radiation therapy is pulmonary toxicity, which leads to various irreversible changes in the lung parenchyma and airways and, as a consequence, to respiratory disorders [5,6]. A significant share of complications in various types of therapy for CLDs are infections, and in the first place-respiratory lesions. Such events are often fatal and determine the life expectancy of patients [7,8].
The presence of concomitant respiratory pathology complicates the process of treatment and recovery of patients with CLDs, as it may be a potential risk factor for an unfavorable prognosis for this category of patients. In addition, this can complicate the diagnostic process, for example, up to 93% of patients with sarcoidosis have such a common syndrome among patients with CLDs as intrathoracic lymphadenopathy [9]. However, this problem is not thoroughly studied today, and there are only a few works on this issue. Thus, according to a study conducted at the Mayo Clinic (USA) and published in 2017 [10], 17% of patients with CLDs had respiratory diseases, and among patients with CLL who died, 23% had respiratory comorbidity. The study previously conducted at our center showed the presence of diagnosed respiratory comorbidity in 9% of patients [11]. Simultaneously, respiratory symptoms were much more common: as an example, shortness of breath had 14.1 % of primary diagnosed with CLDs patients.
According to the Unified Clinical Protocol for the care of patients with lymphoma (2013), chest computed tomography (chest CT) at the stage of initial examination and formulation of clinical diagnosis is mandatory and is carried out to diagnose lymphadenopathy [12], however, diagnosis of lesions of the lung parenchyma and airways is not regulated by protocols. On the other hand, in clinical practice chest CT is not performed in all patients and sometimes is still replaced by chest X-ray. CT is a fundamental method of radiological diagnosis of chest diseases. Its application avoids the necessity for a significant number of invasive diagnostic techniques [13]. The sensitivity of modern CT in the diagnosis of thoracic diseases is over 94% and is the most effective in comparison with other imaging methods [14]. Because the treatment of this category of patients uses drugs that have immunosuppressive and toxic effects, it is important to monitor not only the lymph nodes but also the lung parenchyma, to diagnose infectious complications and the formation of mucositis and pneumofibrosis. Therefore, the problem of determining the nature, extent, and location of lesions of the respiratory system at the stage of initial diagnosis of CLDs is extremely relevant, but today remains open in Ukraine and the world. The study aimed to determine the prevalence, nature, extent, and location of changes diagnosed by high-resolution chest CT in patients with CLDs at the initial examination.

Materials And Methods
Medical records of inpatients who were hospitalized in 2018-2019 in the City Hematology Center of the Municipal Non-Profit Enterprise "City Clinical Hospital № 4" of Dnipro City Council with a confirmed clinical diagnosis of CLD were included in the retrospective study. The results of high-resolution chest CT, which was made during hospitalization using CT (Toshiba Aquilion 160, Japan), were studied. The presence of enlarged lymph nodes (mediastinal paraaortic, paratracheal, sub-carinal, and hilar lymph nodes; and peripheral axillary, cervical and subclavian), nodules, cysts, thickening of the walls of the bronchi, interalveolar septa, areas of ground-glass opacity, consolidation and destruction in the lung parenchyma, pleural effusion, bone-destructive changes were assessed. Statistical analysis was performed in Excel AtteStat 2010 (License number 02260-018-0000106-48794). Quantitative variables were presented as the mean (standard deviation, SD) with the normal distribution of data or as the median (interquartile range) with the abnormal distribution. The distribution of variables was analyzed using the Shapiro-Francia test [15]. Qualitative variables were represented as absolute numbers and percentages (n; %), comparisons of several binary sets were performed using the Chi-square criterion.

Results
A total of 1,004 hospitalized patients with confirmed CLDs were included in the analysis.  . Among patients with CLL, respiratory comorbidity was found in two patients-community-acquired pneumonia in one patient (0.8% of the entire cohort and 3.3% of patients with CLL) and chronic bronchitis in one patient (0.8% of the total cohort and 3.3% of patients with CLL); among patients with lymphoma respiratory pathology was registered in five patients-chronic obstructive pulmonary disease (COPD) in one patient (0.8% of the whole cohort and 3.7% of patients with lymphoma), asthma in one patient (0.8% of the whole cohort and 3.7% of patients with lymphoma), sinusitis in one patient (0.8% of the entire cohort and 3.7% of patients with lymphoma), and mucositis in two patients (1.7% of the entire cohort and 7.4% of patients with lymphoma); among nine patients with MM and respiratory comorbidity, one patient had sinusitis (0.8% of the whole cohort and 1.9% of patients with MM), one patient had tracheitis (0.8% of the whole cohort and 1.9% of patients with MM), two patients with chronic bronchitis (1.7% of the entire cohort and 3.9% of patients with MM), and community-acquired pneumonia in five patients (4.2% of the total cohort and 9.8% of patients with MM). Patients with chronic respiratory comorbidity had a stable phase when were enrolled in the study. Reported community-acquired pneumonia in six cases was present at the time of initial diagnosis of CLDs. Out of 119 patients, 81 patients underwent chest CT examination (68.1%), while chest radiography in two projections was performed in 61 patients (51.3%) at the primary care level. Both types of studies were performed in 23 patients (19.3%).
During analyzing the nature of changes in chest CT in patients with newly diagnosed CLDs ( Table 2), it was found that about a third of the studied cohort of patients had both central and peripheral lymphadenopathy. These symptoms dominated in patients with CLL, lymphoma, and other CLDs, in contrast to patients with MM. In turn, patients with MM had parenchymal lesions statistically significantly more frequently. However, no statistically significant differences were detected when the prevalence of their categories was compared.
Predictably, bone-destructive changes were statistically significantly more common in patients with MM than in other groups of CLDs patients.   Eleven patients (13.6%) had foci of consolidation in the pulmonary parenchyma, the localization is presented in Figure 1. Nodules were detected in three patients (3.7%), had a diameter of up to 3.5 mm, and were localized in the right middle lobe (RML), right lower lobe (RLL), and left upper lobe (LUL). Bronchial wall thickening was determined in 12 patients among newly diagnosed patients with CLDs who underwent chest CT before the start of specific treatment (14.8%). One patient (8.3%) had bronchial wall thickening in all lung lobes, two patients (16.6%) only in LLL, one (8.3%) only in LUL, and one (8.3%) in the lower lobes of both lungs. Pneumofibrosis was determined on chest CT in 14 patients (18.3%), the localization is presented in Figure 2. Among them, one patient (7.1%) had a clinical diagnosis of chronic obstructive pulmonary disease. Lymphadenopathy on chest CT was detected in 31 patients (38.3%), including 13 patients with CLL (41.9%), 13 with lymphoma (41.9%), two patients with MM (6.5%), and three with other CLDs (9.7%). Mediastinal lymph nodes predominated among the enlarged lymph nodes detected on chest CT (33.3%), including paraaortic in nine patients (11.1%), paratracheal in five (6.2%) patients, subcarinal in four (4.9%) patients, and hilar lymph nodes in 10 primary patients with CLDs (12.3%). Peripheral lymphadenopathy occurred in 24 (29.6%) cases: axillary lymph nodes were enlarged in 21 patients (25.9%), cervical lymph nodes in eight patients (9.9%), and subclavian in five patients (6.2%). Chest CT scans of the patients from the study cohort are presented in Figures 3-4.

Discussion
The major strength of this study includes a homogeneous cohort of patients with chronic lymphoproliferative disorders. This is the first study in Ukraine reporting chest CT findings in patients with CLDs, and there are practically no world publications on this topic. Almost half of the patients with CLDs at the initial examination with chest CT revealed signs of pulmonary diseases, which should be taken into account when predicting complications of treatment for such patients. The most common pathological changes detected by chest CT in these patients were lymphadenopathy (about 40% of patients). One-third of patients with CLDs had an increase in mediastinal lymph nodes, including 6.2% of paratracheal lymph nodes, 4.9% of bifurcation, and 12.3% of hilar lymph nodes, which could potentially affect airway patency and pulmonary function. Predictable, lymphadenopathy was more common in patients with CLL and lymphoma, but also was detected in 7% of patients with MM, which is not classical for this disease and needs further diagnosing to detect the possible primary cause of lymphadenopathy in these patients. Among the lesions of the lung parenchyma, pneumofibrosis with predominant localization in the lower lobe of the left lung and foci of consolidation of the lung parenchyma in the lower lobes of both lungs were most often diagnosed.
Some limitations of our study should be acknowledged. First, the monocentric design has an impact on the generalizability of these findings, so data can be extrapolated with caution to the entire patient population. Second, the study was retrospective, so not everyone patient had a chest CT scan. At the stage of initial examination and formulation of clinical diagnosis, chest CT was performed only in 68% of patients. Finally, data on comorbidity were collected only from medical records, consequently, we did not have an opportunity to exclude the likeliness of comorbidity underdiagnosing. Future studies will have a prospective design that will allow us to assess the real prevalence of respiratory comorbidity in patients with CLDs (using physical examination, anamnesis data, chest CT in all patients, and spirometry), as well as its relationship with the outcome of the underlying disease.