Fayoum experience in the ultrasonographic evaluation of diffuse parenchymal lung disease

Objective Chest ultrasound has many uses, both diagnostic and interventional. It may be used for the diagnosis of multiple pleural diseases (pleural effusion, pleural masses, and pneumothorax). It is also used in the diagnosis of diseases caused by lung parenchymal lesions, such as neoplasms, pulmonary embolism, pneumonia, and lung abscesses. Aim This study aimed to evaluate the sonographic features of diffuse parenchymal lung disease (DPLD). Design This was a prospective study. Setting The study was carried out at Fayoum University Hospital in Egypt during the period spanning from January 2017 to June 2017. Patients and methods This study included 120 participants. Sixty of them were diagnosed as having DPLD. For the diagnosis of these cases, we need a full medical history, a detailed clinical examination, spirometry, 6 min walk test, arterial blood gases’ analysis, high-resolution computed tomography, and chest ultrasound. The other 60 were studied as controls. Statistical analysis The data were collected and coded to facilitate data manipulation, and they were double entered into Microsoft Access; the data analysis was performed by using SPSS software, version 18 in Windows 7. The simple descriptive analysis was carried out in the form of numbers and percentages for qualitative data and arithmetic means as central tendency measurement, SDs as measure of dispersion for the quantitative parametric data and the inferential statistic test. Results There was a female predominance with a wide range of age. Most of the cases were nonsmokers, breeding birds, and exposed to biomass. All cases had diffuse bilateral B-lines. There was a negative relation between the Warrick score on the one hand and the B-line number, PaO2, 6 min walk test, and forced vital capacity on the other hand. In contrast, a positive relation was demonstrated between the Warrick score and B-line distance and pleural thickness. Most of the studied patients (71.6%) had irregular and thickened pleura and (51.6%) had an abolished lung sliding. Conclusion Chest ultrasound has a significant role in the diagnosis of DPLD and also in estimating the severity of the disease according to the number and the distance between B-lines. Multiple B-lines in the combination of thickened and irregular pleural line are highly suggestive of DPLD.


Introduction
Diffuse parenchymal lung disease (DPLD) is a group of nonmalignant and noninfectious processes caused by multiple causative agents that may be known or unknown [1].
High-resolution computed tomography (HRCT) may be considered as the gold standard technique for the diagnosis of DPLD, and there are also other noninvasive and invasive technique for diagnosis of DPLD, such as laboratory and serological tests, pulmonary function tests, transthoracic ultrasound, bronchoscopy with a bronchoalveolar lavage and the transbronchial lung biopsy [2].
Chest ultrasonography can play an important complementary role, particularly in the critical care setting where conventional radiography is often suboptimal.
Its lack of ionizing radiation, bedside availability, and dynamic imaging capacity afford ultrasound certain advantages over other techniques [3].

Aim of this study
This study aimed to evaluate the sonographic findings of DPLD.

Patients and methods
A case-control prospective study was carried out in the Chest Department, Fayoum University, during the period spanning from January 2017 to June 2017. The study included 120 participants: 60 patients who were diagnosed as having DPLD and 60 controls from the Fayoum governorate. The patients were diagnosed according to the diagnostic algorithm of DPLD [4].
The study protocol was approved by the Ethical Committee of Faculty of Medicine in Fayoum University, and informed consent was obtained from all patients.

Exclusion criteria
The cases with pulmonary edema of different causes, pneumonia, pulmonary embolism, and atelectasis were excluded from this study because B-lines may be detected in these cases [5].
All participants were submitted for the following: The values were obtained when the patient exerted his maximum effort to avoid any expected error in diagnosis. The absolute values and the percentages of spirometric parameters predicted from the participant's age, sex, and height were calculated. Restrictive ventilatory defect was defined by spirometric findings of FEV1/FVC greater than 70% predicted and FVC less than 80% predicted [6].

6-min walk test
It is measurement of the distance that patients can quickly walk on a flat, hard surface in a period of 6 min. Reference values are 576 m for healthy male patients and 494 m for healthy female patients [7]. Before and after the test, oxygen saturation % was measured by a pulse oximeter, and desaturation was defined as a fall in oxygen saturation to 88% or less during the 6 min walk test [8].

High-resolution computed tomography
The device used was Asteion Toshiba four-slice CT scanner (Toshiba, Canon Medical System, California, USA).
In this study, pulmonary involvement is identified and scored according to the semiquantitative modified Warrick score [9].
Modified Warrick scoring was obtained by summation of the point values of parenchymal abnormality pattern (the severity score), which range from 0 (normal) to 15 (all lesions are present), and the extension score was obtained by localization of the number of pulmonary segments involved in both lungs, which ranged from 0 normal to 15, at which more than nine pulmonary segments are involved.
Thus, the results of summation of the total Warrick score ranged from 0 no involvement to 30, which represent the worst involvement.
The results were expressed as follows: (1) Normal: 0 point.

Chest ultrasound
Chest ultrasound was performed for all participants using Philips ClearVue 320 (Philips, Amsterdam, Netherlands) device.
Lung ultrasound was performed using the low frequency (2.5-5 MHz) convex probe for lung examination and the high frequency (7.5-10 MHz) linear probe for pleural examination.
There are different positions during examination: Supine (ventral images), sitting (dorsal and lateral images), and raising the arms and crossing them behind the head, leading to extension of the intercostal spaces and facilitating access.
Assessment of the following: (1) B-lines: The predominant bilateral B-lines (vertical hyperechoic lines) arising from the pleural line that spread towards the edge of the screen without fading, causing erasing of A-lines and move with the lung sliding [10]. A distance between two adjacent B-lines close to the pleural line is measured and expressed in milliliters [11]. A positive region is defined through the presence of three or more B-lines in the longitudinal plane between two ribs. The positive examination is defined through the presence of two or more positive regions on both sides. The number of B-lines is counted per zone [12]. (2) The lung sliding (to-and-fro dynamic movement of the lung during respiration, which was visible at the pleural line). (3) Assessment of pleura as regards pleural thickness and pleural surface (either smooth, irregular). The pleural thickening is known as focal or diffuse echogenic lesions arising from visceral or parietal pleura and the width of which is greater than 3 mm with or without the irregular pleural surface [13].

Statistical analysis
The data were collected and coded to facilitate the data manipulation and double entered into Microsoft Access, and the data analysis was performed by using SPSS software (SPSS Inc. Released 2009. PASW Statistics for Windows, Version 18.0. Chicago: SPSS Inc).
The simple descriptive analysis was carried out in the form of numbers and percentages for qualitative data, and arithmetic means as central tendency measurement, SDs as measure of dispersion for quantitative parametric data and inferential statistic test.

In cases of quantitative parametric data
The dependent Student's t test was used to compare measures of two independent groups of quantitative data.
One way analysis of variance test was used to compare more than two independent groups of quantitative data.
For qualitative data χ 2 test was used to compare two or more than two qualitative groups.
The P value less than 0.05 was considered the cut-off value for significance.

Results
This study included 60 patients, of whom 54 (90%) were female patients and six (10%) were male patients; the age group ranged from 20 to 75 years, with a mean of 47.5±13.6 years.
Thirty percent of cases presented with dyspnea only, 36.7% presented with dyspnea and cough, and, finally, 33.3% of them presented with dyspnea, cough, and wheezing.
Most of the patients were exposed to biomass and breeding birds (80%), and 90% of patients were nonsmokers (41.6%, n=25) having clubbing.
These data are clarified in Table 1, which represents the demographic data of the studied patients.
By HRCT, most patients (50%, n=30) were diagnosed as having hypersensitivity pneumonitis and 16.6% (n=10) were diagnosed as having idiopathic pulmonary fibrosis. This is clarified by Table 2.
The mean and SD of B-lines' number was 6.3±2.1. The B-line distance ranged from 3 to 12 mm with mean ±SD of 7.8±2.7.
The range of the pleural thickness in all studied patients was from 2 to 19 mm with mean±SD of 9.1 ±5.9.
There was an irregular pleural line in 71.6% (n=43), whereas 28.3% (n=17) had a regular pleural line. All these data are clarified in Table 3.
The pulmonary involvement is scored according to the semiquantitative HRCT scoring system (modified Warrick score), which is shown in Table 4.
According to this score, the pulmonary involvement was classified into three groups: These data are illustrated in Table 5. Table 6 illustrates that there is a significant positive relation with P value less than 0.05 between Warrick score and the distance between B-lines; which indicated that an increase in Warrick score was associated with an increase in distance between B-lines. Moreover, there is a significant negative relation between Warrick score and the following: number of B-lines, PaO 2 , 6 min walk test, FVC, and pleural line thickness, which indicated that an increase in Warrick score was associated with a decrease in all these variables.
In contrast, there was no relation between Warrick score and duration of symptoms. Figures 1-6 show these relations. Table 7 illustrates that there is a significant difference between the different degrees of Warrick score according to regularity, plural line thickness, and B-line distance, with a high percentage of irregular and thickened plural line and wide distance being noticed among severe degree followed by moderate degree cases.
There was a statistically significant difference between the different degrees of Warricks score according to Bline numbers, 6 min walk test, and FVC, with a high mean among mild degree of severity cases. This is clarified by Figs 7 and 8.

Discussion
It has been highlighted that chest ultrasound is sensitive to variations of pulmonary content and balance between air and fluids. In the normal lung, the ultrasound waves are completely reflected by the air, whereas, in the lung diseases that impair alveolar air content and increase the interstitial and alveolar fluids, a particular artifact will be created [14].
The ultrasound evaluation of DPLD is determined by the presence and quantification of B-lines, they are generated by reflection of the ultrasound beams from the thickened subpleural interlobar septa at the lung surface interface [15].  Pleural thickness (mm) in total studied patients (mean±SD) 9.1±5.9 Pleural thickness for the group with normal pleural thickness (mean±SD) 2.7±0.33 Pleural thickness for the group with thickened pleural line (mean±SD) 11 Figure 2 Illustrates the relation between Warrick score and B-line distance.

Figure 1
Illustrates the relation between Warrick score and B-line number.
In this study, 30% of cases presented with dyspnea only, 36.7% presented with dyspnea and cough, and, finally, 33.3% of them presented with dyspnea, cough, and wheezing. The studied group showed a female predominance with a wide range of age (20-75). Most of them were breeding birds and exposed to biomass (80%, n=48), and 90% were nonsmokers.
By using transthoracic ultrasound, all patients had diffuse and more than three B-lines bilaterally. The best time to see these B-lines is during the real-time examination, and this seems to be due to some lines being less clarified on frozen sonograms.
In this study, there are at least four B-lines per scan. Mean and SD of the B-lines' number was 6.3±2.1, whereas the distance between two adjacent B-lines ranged from 3-12 mm, and mean±SD was 7.8±2.7. This is illustrated in Table 3.
In this study, pulmonary involvement is scored according to semiquantitative HRCT score systems (modified Warrick score), which is shown in Table 4.
According to this score, the pulmonary involvement was classified into three groups: (1) Mild group (26.7%, n=16) with a mean±SD of 6.1 ± 0.9. Illustrates the relation between Warrick score and PaO 2 .

Figure 4
Illustrates the relation between Warrick score and 6MWT. 6MWT, 6 min walk test.

Figure 5
Illustrates the relation between Warrick score and FVC. FVC, forced vital capacity.

Figure 6
Illustrates the relation between Warrick score and pleural thickness. These data are illustrated in Table 5.
The same results were obtained by Mohammadi et al. [16]; they found a relation between the Warrick score and the severity of pulmonary involvement (r=0.695, P<0.001).
In this study, there was a significant negative relation, with P value less than 0.001, between the Warrick score and number of B-lines (Fig. 1).
However, there is a significant positive relation, with P value less than 0.05, between Warrick score and Blines' distance (Fig. 2).
These data are shown in Table 6.
These results were in agreement with Hasan and Makhlouf [17], who evaluated 61 patients with DPLD, as they concluded that the B-lines seem numerous, with a narrow distance between them in those with early DPLD.
They illustrated that there is a relation between the B-line distance and the severity of the disease on chest HRCT at which B3 (the HS, highly significant. The bold means the highest values and the bold of P value means high significant.

Figure 7
Illustrates the regularity of pleural line between different severity degrees among cases.
Also these results are incompatible with the results of Farag et al. [18] who found that the degree of interstitial affection on chest HRCT, which was detected by the modified Warrick score, is inversely HRCT chest of a patient with HP showed ground-glass opacity. Ultrasound of the same patient showed diffuse B-lines with a narrow distance between them (about 3 mm). HP, hypersensitivity pneumonitis; HRCT, high-resolution computed tomography.

Figure 8
Illustrates the thickness of the pleural line among different severity degrees among cases.

Figure 10
HRCT of the chest showed subpleural reticulation with honey combing. Ultrasound of the same patient showed diffuse B-lines with a wide distance between them (about 7 mm). HRCT, high-resolution computed tomography.
correlated with the total number of B-lines among mild, moderate, and severe groups.
In the current study, there is a significant negative correlation between Warrick score and the following (and they are): PaO 2 , 6 min walk test, and FVC% (Figs 3-5), which indicates that the well-established pulmonary fibrosis in HRCT may be associated with more impairment in pulmonary function test (PFTS).
Hence, the severe group has the lowest mean of FVC% (45±9.4), and this is shown in Table 7.
According to the pleural line, the linear probe was used for more details.
In this study, the range of the pleural thickness in all studied patients was from 2 to 19 mm with mean±SD of 9.1±5.9 as shown in Table 3.
There was an irregular pleural line in 71.6% (n=43), whereas 28.3% (n=17) had a regular pleural line. All these data are clarified in Table 3.
The mild group had pleural thickness with a mean±SD of 3.8±1.6, the moderate group had pleural thickness with a mean±SD of 6.4±3.4, and the severe group had pleural thickness with a mean±SD of 16.1±2.3, and this is clarified in Table 7.
Hence, there is a positive correlation between the Warrick score and pleural thickness (Fig. 6).
This was compatible with the results of Farag et al. [18] who found that there was an important significant difference of pleural thickness in millimeter (mean ±SD) between all groups involved.
In this study, 43.8% of the mild group had irregular thickened pleural line, 65.2% of the moderate group had irregular thickened pleura line, whereas 100% of severe cases showed irregular thickened pleural line (Figs 7 and 8).
These data are illustrated in Table 7.
These results were supported by the study of Farag et al. [18], as they demonstrated that 44.4% of mild cases had irregular thickened pleural line just at the lung base, whereas all severe cases showed irregular pleural line over the whole lung (100%), thickened pleura (83.3%), and the moderate group showed less value for irregularity (83.3%) and thickening (54.2%). Table 7 illustrates that the lung sliding was present in the mild group, whereas it is abolished in severe cases of parenchymal affection in HRCT.
This result was in agreement with Agmy et al. [19] who found that abolished lung sliding had a highly significant positive correlation with reticular opacities on multidetector computed tomography (MDCT), and this could be attributed to the fact that fibrosis, which affects the visceral pleura, will be followed by restriction of lung ventilation and expansion, leading to impairment of lung sliding.

Limitations of the study
The diffusion capacity of carbon monoxide and total lung capacity were not measured, and the number of patients (60) is small, because the study was controlled by a specific duration (January 2017 to June 2017).

Conclusion
Chest ultrasound has an important role for the diagnosis and evaluation of DPLD patients. Numerous B-lines that are distributed over the lung surface constitute the most important sonographic feature in combination with a thickened irregular pleural line and the abolished lung sliding.
Chest ultrasound may give an idea about early or advanced DPLD through the number of B-lines and the B-line distance.

Financial support and sponsorship
Nil.

Conflicts of interest
There are no conflicts of interest.