Acute eosinophilic pneumonia: radiologic and clinical features

PURPOSE

To determine the radiographic and computed tomographic (CT) characteristics of acute eosinophilic pneumonia.

MATERIALS AND METHODS

Twelve consecutive patients with acute eosinophilic pneumonia were included in the study. The diagnosis was based on clinical symptoms and results of bronchoalveolar lavage. Plain chest radiographs were obtained in all patients; CT scans were obtained in three patients. Two thoracic radiologists reviewed the radiographs and CT scans.

RESULTS

Ten patients had bilateral areas of air-space opacity on images obtained at presentation; in seven of these patients, interstitial areas of opacity were also present. Two patients had bilateral interstitial areas of opacity and no areas of air-space opacity. Interlobular septal thickening and ground-glass attenuation were present on CT scans in two patients; patchy bilateral consolidation was present on CT scans in one patient. Pleural effusion was present on radiographs in seven patients (58%) and was bilateral in five. Pleural effusion was present at some point during the course of disease in all patients. In all patients, air-space disease markedly improved within 3 days of initiation of treatment with corticosteroids.

CONCLUSION

Acute eosinophilic pneumonia should be considered as a possible diagnosis when a previously healthy person presents with acute respiratory failure of unknown origin.

Obliterative bronchiolitis: individual CT signs of small airways disease and functional correlation

PURPOSE

Individual features of small airways disease depicted at computed tomography (CT) were correlated with functional indexes in patients with obliterative bronchiolitis.

MATERIALS AND METHODS

Fifteen patients (all women) who fulfilled the strict criteria for diagnosis of obliterative bronchiolitis underwent thin-section CT at full inspiration and at end expiration. The CT scans were scored by two observers for extent of decreased attenuation of the lung parenchyma; end-expiration CT signs of air trapping; and bronchial dilation, wall thickening, and mucous plugging. The functional importance of each CT sign was evaluated.

RESULTS

Areas of decreased attenuation were present in all patients (median score at end expiration, 61%; range, 21%-83%). Bronchial wall thickening was identified in 13 of the 15 patients. Correlations of the extent of decreased attenuation and measures of air-flow obstruction were strongest between decreased attenuation at end expiration and air flow at low lung volumes (r(s) = -.70, P < .005). This relationship remained intact after correction for the severity of bronchial wall thickening.

CONCLUSION

In patients with obliterative bronchiolitis, the extent of decreased attenuation at CT was most strongly related to depression of pulmonary function tests of the small airways. Decreased attenuation is the cardinal sign for further quantitative studies of obliterative bronchiolitis.

Pulmonary alveolar microlithiasis. A clinicopathologic and chemical analysis of seven cases

OBJECTIVE

To determine the clinical features and outcome of patients with pulmonary alveolar microlithiasis and to determine the chemical composition of the microliths.

CASE MATERIAL

We studied seven cases of pulmonary alveolar microlithiasis. The patients were six women and one man, aged 19 to 70 years (mean age 44.5 years). Clinically, five patients were known to have suffered from this condition for 5 to 41 years. One patient presented with shortness of breath, and another had a gradual decrease in exercise tolerance. None of the patients had a previous history of disturbances in metabolism or any other relevant medical condition. Reports on radiographic studies were available in six cases, and chest radiographs were available for review in the seventh case. They all showed diffuse bilateral pulmonary infiltrates. Open lung biopsies were performed in two patients, and autopsy lung material was reviewed in five patients.

RESULTS

Histologically, the lung showed the typical features of pulmonary alveolar microlithiasis, that is, presence of numerous microliths filling the alveolar spaces with either a normal or thickened fibrotic interstitium. Chemical analysis performed on the lung tissue of six of these patients revealed that the microliths consisted principally of calcium and phosphorus salts. Five of these patients died of respiratory failure; however, their deaths occurred from 5 to 41 years following the initial diagnosis. No follow-up information was obtained in two patients.

CONCLUSIONS

The findings of this study confirm that pulmonary alveolar microlithiasis can be seen in any age group and that the microliths are composed principally of salts of calcium and phosphorus. Additionally, these cases confirm that the disease typically follows a protracted course.

Lung epithelial permeability and bronchial responsiveness in subjects with stable asthma

Lung epithelial permeability of asthmatic patients has been reported to be similar or lower than that of healthy subjects and to be correlated or not to bronchial hyperresponsiveness. To clarify these discrepancies, we evaluated 99mTc-DTPA pulmonary clearance in a group of carefully selected asthmatic patients with mild, stable asthma (n = 13; seven women; mean age +/- SD = 27.69 +/- 6.63 years), and compared them with a group of healthy, nonsmoking subjects (n = 8; six women; mean age +/- SD = 24.38 +/- 5.15 years). Selection criteria for asthmatics were as follows: baseline FEV1 > or = 80% of predicted values, no bronchial infections, and/or no asthma attacks during 4 weeks prior to study and peak expiratory flow rate variability lower than 20%, over a period of 3 weeks. Patients controlled symptoms with beta 2-adrenergic drugs only, regularly or on demand. Mean baseline FEV1 (+/-SD) as percent of predicted was 102.38 +/- 13.97 and 112.88 +/- 18.36, respectively (p < 0.05). In the asthmatic group, bronchial responsiveness to methacholine (PC20 M FEV1) ranged between 0.55 and 28.5 mg/mL. Mean value (+/-SD) of DTPA clearance from lungs to blood (evaluated on the first 10 min out of 30 min of the curves) in the asthmatic group was not different from that of control group (68.31 +/- 21.46 and 69.5 +/- 15.73). In the asthmatic patients, there was no correlation between PC20 M values and DTPA T1/2 min of the whole lung, nor between PC20 M and inner and outer lung clearance zones. Moreover, both in asthmatics and healthy subjects, DTPA clearance of outer (alveolar) zones was significantly faster than that of inner (bronchial) zones (57.69 +/- 19.94 vs 102.08 +/- 38.19, p < 0.001, and 59.75 +/- 12.49 vs 103.5 +/- 31.86, p < 0.003, respectively). Our data show that DTPA clearance in patients with stable asthma is similar to that found in healthy subjects; it is not correlated to degree of bronchial responsiveness and occurs more rapidly in the outer zones than in the inner zones, both in asthmatic patients and in healthy subjects. Thus, to date, DTPA clearance index is not a valid tool for identifying and/or monitoring asthmatic patients.

Cryptogenic fibrosing alveolitis and the fibrosing alveolitis of systemic sclerosis: morphological differences on computed tomographic scans

BACKGROUND

The purpose of this study was to identify morphological differences on the computed tomographic (CT) scan between cryptogenic fibrosing alveolitis (CFA) and the fibrosing alveolitis associated with systemic sclerosis (FASSc), and to examine their biological relevance.

METHODS

One hundred and seven patients with CFA (n = 55) or FASSc (n = 52) who had undergone thin section CT scanning were included. Multivariate analysis was used to identify morphological differences on the CT scans between lone CFA and FASSc, and to determine whether the pattern and distribution of disease on the CT scans were functionally significant (as judged by the lung transfer factor (TLCO), forced vital capacity (FVC), and arterial oxygen tension (PaO2)) or predictive of survival (independent of the type and extent of fibrosing alveolitis, age, sex, and smoking history).

RESULTS

Increasingly extensive disease on CT scans was associated with a coarser reticular pattern (increase in reticular score per percentage increase in disease extent = 0.06, 95% confidence interval (CI) 0.03 to 0.09, p < 0.0005) and increasing upper zone involvement (increase in ratio of upper zone to total disease per percentage increase in disease extent = 0.002, 95% CI 0.000 to 0.003, p < 0.04). Patients with CFA were characterised by a higher upper zone ratio (difference = 0.08, 95% CI 0.02 to 0.13, p < 0.004) and a weak trend towards a coarser reticular pattern (p = 0.09), independent of disease extent. Smokers with CFA had more upper zone involvement (difference = 0.11, 95% CI 0.05 to 0.16, p < 0.0005) and a coarser reticular pattern (difference in reticular score = 1.92, 95% CI 0.27 to 3.55, p < 0.02) than smokers with FASSc. The extent of disease on the CT scan was predictive of lung function impairment and survival but the pattern and distribution of disease were not.

CONCLUSIONS

Patients with CFA have relatively more upper zone involvement than those with FASSc independent of the extent of disease on the CT scan. This finding may result from smoking related damage but is not functionally significant.

Radiographic patterns of pulmonary disease

Pulmonary radiographs are essential adjuncts to the evaluation and diagnosis of suspected pulmonary disease. In the intensive care unit, radiographs are useful to confirm correct positioning of diagnostic and therapeutic devices. Patterns seen on the radiograph may be within broadly normal limits or may be interpreted as abnormal, especially when placed in the clinical context of a specific patient's problem. The description abnormal can be related to both nonspecific and specific radiographic patterns of disease. Nonspecific radiographic patterns of disease include location of disease, temporal course of disease, pleural abnormalities, hyperinflation, extra-alveolar air, atelectasis, bronchiectasis, and vascular disease. Specific radiographic patterns of disease are discrete anatomic structures seen on a radiograph, for example, cavitary and cystic disease. The interpretation of nonspecific and specific radiographic patterns is useful in diagnosis, selection of treatment, and monitoring of the course of disease and the patient's response to treatment.

Age-related reduction in 99Tcm-DTPA alveolar-capillary clearance in normal humans

Age is known to reduce the efficacy of body organs and systems, even in the absence of disease. The alveolar-capillary clearance (ACC) rate is representative of the alveolar-capillary barrier's functional state. We studied 29 healthy non-smokers, who were selected after clinical and radiographic evaluation. The patients were divided into three groups based on age: Group I, < or = 30 years (n = 10); Group II, 31-55 years (n = 9); Group III, > or = 56 years (n = 10). Each patient inhaled 750 MBq 99Tcm-diethylenetriamine pentaacetate (99Tcm-DTPA) aerosol generated by a Venticis nebulizer (particles with a mean diameter of 1.1 microns) for 5 min. Forty frames of 30 s duration each were acquired and the ACC rates for the right and left lungs determined using a computer program. The mean ACC rates for the three groups were as follows: Group I, 1.31% min-1; Group II, 1.08% min-1; Group III, 0.76% min-1. The differences between Groups I and III (P < 0.001) and Groups II and III (P = 0.03) were shown to be significant. There was no significant difference between Groups I and II. Possible explanations for an age-related reduction in ACC rates include a reduction in the internal alveolar surface, the closure of the small airways, a reduction in the lung blood capillaries and a reduction in cardiac output. We conclude that there appears to be an age-related reduction in ACC rates in healthy non-smokers, even in the absence of clinically and radiographically detectable lung disease. However, larger studies are required.

Alveolar overdistension is an important mechanism of persistent lung damage following severe protracted ARDS

It is now widely accepted that mechanical ventilation may damage the lung, but the mechanism of lung damage is not clear. Possible causes include overdistension of aerated alveoli by inappropriately large tidal volumes (volutrauma), shear stresses generated during the recruitment and de-recruitment of lung units at the junction of aerated and collapsed lung, and infective or ischaemic necrosis of persistently collapsed lung. Computerized tomography allows noninvasive assessment of lung structure during and after acute lung injury, and may provide insight into the mechanism of lung damage. Using serial high resolution computed tomography we documented lung structure one month after recovery from severe protracted adult respiratory distress syndrome (ARDS) in three patients who required mechanical ventilation for between 86 and 97 days; the computed tomograms were repeated at between 5 and 14 months. All three patients had persistent abnormalities of lung structure which were most marked in the anterior regions of the lung. These findings suggest that overdistension of non-dependent lung regions in the main mechanism of lung damage persisting after recovery from severe protracted ARDS.

Diffuse alveolar hemorrhage in limited cutaneous systemic sclerosis with positive perinuclear antineutrophil cytoplasmic antibodies

A 54-year-old woman with limited cutaneous systemic sclerosis (SSc) developed repeated diffuse alveolar hemorrhage with positive perinuclear antineutrophil cytoplasmic antibodies (pANCA). Reports of diffuse alveolar hemorrhage in SSc have been rarely described only in the diffuse clinical variant. To our knowledge, this is the first report of pANCA positive diffuse alveolar hemorrhage in the limited form of SSc. Diffuse alveolar hemorrhage is a rare, but potentially life threatening cause of respiratory distress in the limited form of SSc that should be recognized and treated effectively.

Gallium-67 scanning for detection of alveolitis in idiopathic pulmonary fibrosis and sarcoidosis

OBJECTIVE

To investigate 67Gallium scanning for the detection of alveolitis in idiopathic pulmonary fibrosis and sarcoidosis.

PATIENTS AND METHODS

67Ga scintigraphy was performed in 14 patients with IPF (age: mean +/- s = 58.3 +/- 10.3 years), 13 patients with sarcoidosis (age: mean +/- s = 46.9 +/- 9.0 years) and 11 controls (age: mean +/- s = 44.0 +/- 10.5 years).

RESULTS

Lung/thigh ratio was respectively 3.18 +/- 0.07 in IPF, 3.12 +/- 0.94 in sarcoidosis and 2.11 +/- 0.26 in controls. IPF and sarcoidosis groups had a significantly higher lung/thigh ratio comparing with the control group (P < 0.05). Three patients received 67Ga scanning examination after corticosteroid therapy and the 67Ga uptake decreased.

CONCLUSIONS

The 67Ga scintigraphy is useful in monitoring the activity and extent of alveolitis.

Abnormalities of the airways and lung parenchyma in asthmatics: CT observations in 50 patients and inter- and intraobserver variability

The purpose of the study was to evaluate the CT abnormalities of airways and lung parenchyma in asthmatic patients and to assess inter- and intraobserver variability for these abnormalities. The CT scans of 50 asthmatic patients and 10 healthy volunteers were assessed independently by four independent chest radiologists who were masked with respect to the clinical information. Bronchiectasis involving mostly subsegmental and distal bronchi was noted in 28.5% of the asthmatic subjects and none of the non-asthmatics. Bronchial wall thickening, small centrilobular opacities and decreased lung attenuation were observed in 82%, 21% and 31% of asthmatic patients respectively, compared with 7%, 5% and 7% of healthy subjects. The intra- and inter-observer agreements for these four CT abnormalities were measured by the kappa statistic and ranged from 0.60 to 0.79 and from 0.40 to 0.64, respectively. It is concluded that asthmatic patients may exhibit bronchial wall thickening, bronchiectasis and morphological abnormalities suggestive of distal airways disease that can be assessed on CT scans with a clinically acceptable observer variability.

The relationships between the radionuclide alveolar integrity study and the pulmonary function test

This study evaluated the relationships between the alveolar integrity and the pulmonary function in 33 male subjects, including 18 COPD patients, 10 normal volunteers and 5 NIDDM patients. Alveolar integrity was evaluated by dynamic Tc-99m DTPA/HMPAO radioaerosol lung scintigraphies. The speed of radioaerosol clearance is represented as a slope from the lungs to the blood stream. Comprehensive pulmonary function tests included forced expiratory volume, flow-volume loop, and diffusion capacity of a single breath CO method. The results showed the following: that (1) there are no correlations between the slopes and the parameters of pulmonary function tests, and (2) that the clearance of DTPA aerosols is faster than the clearance of HMPAO, and that these differences are significant. In conclusion, the radionuclide alveolar integrity study should be considered as a new method for evaluating lung function, and should be differentiated from traditional pulmonary function tests, and, secondly, lipophilic HMPAO aerosols are slower to clear than those of hydrophilic DTPA, which suggests at least two different clearance mechanisms for radioaerosols.

Radiological diagnosis of pneumonia in children

During 12 months in 1981-82, 201 children were hospitalized due to radiologically verified definite or probable pneumonia. In 1985, 194 chest radiographs (anteroposterior views) were re-evaluated jointly by two radiologists, and classified into three categories: alveolar, interstitial and probable pneumonia. In 127 cases definite pneumonia was diagnosed on both occasions, alveolar in 48 cases and interstitial in 79 cases. Variation between the two evaluations 3 years apart was observed in 46 (24%) of the 194 cases; the adjusted kappa (0.47) was in the modest region. Factors contributing to this variation were young age, less than 12 months, and the presence of interstitial infiltration, bronchial obstruction and low C-reactive protein. Factors associated with less marked variation were the presence of alveolar infiltration, auscultatory fine rates and elevated C-reactive protein. The microbial aetiology of infection, assessed by viral and bacterial antigen and antibody assays, showed no association with diagnostic variation. A lateral view of the chest radiograph was obtained from 158 patients; it was positive in 99 (91%) of the 109 cases with definite pneumonia. In only three cases the diagnosis was based on the lateral view alone. Our results show that the radiological diagnosis of pneumonia is difficult in children, especially in young children with interstitial pneumonia.

Uncommon pneumoconioses: CT and pathologic findings

PURPOSE

To correlate the computed tomographic (CT) features of pneumoconioses with histologic findings.

MATERIALS AND METHODS

Thin-section CT scans obtained in 48 patients with a history of occupational exposure to dust and radiographic changes suggestive of pneumoconiosis were retrospectively reviewed. Histologic samples were available in 22 cases.

RESULTS

The most common CT features were as follows: in 21 arc welders, ill-defined micronodules concentrated in the centrilobular regions (n = 15); in 19 graphite workers, small nodular hyperattenuating areas (n = 17) (ill defined or well defined, corresponding to macular lesions along the walls of bronchioles and nodules, respectively), interlobular septal thickening (n = 11), and large hyperattenuating areas (n = 10); in aluminum pneumoconiosis, predominant reticular (n = 2), nodular (n = 2), and upper-lobe fibrosis (n = 2); and in hard-metal pneumoconiosis, multilobular ground-glass attenuation and consolidation with shrinkage (corresponding to marked intra-alveolar desquamation and multinucleated giant cells with mural mononuclear cell infiltrate).

CONCLUSION

Predominant findings are characteristic in each type of pneumoconiosis and are depicted at thin-section CT.

The use of 99Tcm-Technegas in the investigation of patients with pulmonary thromboembolism

Pulmonary embolism remains a major cause of morbidity and mortality, being responsible for an estimated 200,000 deaths per annum in the USA and 21,000 per annum in the UK. Lung scintigraphy is in many instances the investigation of choice in suspected pulmonary thromboembolism. A normal perfusion lung scan excludes pulmonary embolism. An abnormal perfusion scan, while being sensitive, is of low specificity for the diagnosis of pulmonary embolism and needs to be complemented with a ventilation study. Lung ventilation has been studied using inert gases of radiolabelled aerosols. The new radiopharmaceutical 99Tcm-Technegas is a suspension of 99Tcm-labelled, ultrafine, carbon particles produced in an atmosphere of high-purity argon. The size of the particles is of the order of 0.005-0.2 microns, which assures good peripheral penetration and alveolar deposition. After inhalation, static images in multiple projections may be acquired. The resultant images are of excellent technical quality and several publications have shown the value of Technegas images as an adjunct to perfusion imaging in suspected pulmonary thromboembolism. Studies comparing Technegas images with other ventilatory radiopharmaceuticals have in the majority of instances reported comparable diagnostic qualities.

Image processing in nuclear pneumology

The general aspects of methodology and image processing used in some of the nuclear pneumology techniques developed by our group are briefly described in this presentation. Three main subjects are considered: 1--Studies on pulmonary circulation, 2--Studies on pulmonary ventilation and perfusion and 3--Studies on alveolar-capillary barrier permeability. This is a multidisciplinary work that is the joint effort of three department: Biophysics Service (Faculty of Medicine, University of Coimbra), Department of Electronics and Telecommunications (University of Aveiro) and Pneumology Center (University of Coimbra).

Magnitude and time course of acute hypoxic pulmonary vasoconstriction in man

Acute hypoxic pulmonary vasoconstriction has an established role in the preservation of ventilation-perfusion balance. To further characterize this homeostatic response in man we have attempted to measure both the time course and magnitude of blood flow diversion from single hypoxic lobes. Lobar hypoxia (mean PO2 38 +/- 1.5(SEM)mmHg, mean PCO2 39.9 +/- 0.9 mmHg) was induced by inflating catheter-tip balloons in left upper lobe bronchi during fibreoptic bronchoscopy under local anaesthesia in 8 normal subjects. An index of lobar blood flow was obtained by acquiring dynamic scintigraphic lung images during a continuous intravenous infusion of the short-lived radioisotope krypton-81m dissolved in 5% glucose solution. In 3 subjects blood flow to the occluded lobes was monitored while the lobes were maintained under hyperoxic conditions (mean PO2 127.8 +/- 31.5 mmHg, mean PCO2 40.2 +/- 1.3 mmHg). Under hypoxic conditions the blood flow to the occluded lobes fell to 53% of baseline after 5 min with a mean time constant of 151 +/- 24.8 sec. Under hyperoxic conditions there was no significant change from baseline blood flow. We conclude that this technique has allowed us to monitor both the dynamic and steady state responses of the pulmonary circulation to lobar hypoxia in man.

Different ventilatory approaches to keep the lung open

OBJECTIVES

To study the ability of different ventilatory approaches to keep the lung open.

DESIGN

Different ventilatory patterns were applied in surfactant deficient lungs with PEEP set to achieve pre-lavage PaO2.

SETTING

Experimental laboratory of a University Department of Anaesthesiology and Intensive Care.

ANIMALS

15 anaesthetised piglets.

INTERVENTIONS

One volume-controlled mode (L-IPPV201:1.5) and two pressure-controlled modes at 20 breaths per minute (bpm) and I:E ratios of 2:1 and 1.5:1 (L-PRVC202:1 and L-PRVC201.5:1), and two pressure-controlled modes at 60 bpm and I:E of 1:1 and 1:1.5 (L-PRVC601:1 and L-PRVC601:1.5) were investigated. The pressure-controlled modes were applied using "Pressure-Regulated Volume-Controlled Ventilation" (PRVC).

MEASUREMENTS AND RESULTS

Gas exchange, airway pressures, hemodynamics, FRC and intrathoracic fluid volumes were measured. Gas exchange was the same for all modes. FRC was 30% higher with all post-lavage settings. By reducing inspiratory time MPAW decreased from 25 cmH2O by 3 cmH2O with L-PRVC201.5:1 and L-PRVC601:1.5. End-inspiratory airway pressure was 29 cmH2O with L-PRVC201.5:1 and 40 cmH2O with L-IPPV201:1.5, while the other modes displayed intermediate values. End-inspiratory lung volume was 65 ml/kg with L-IPPV201:1.5, but it was reduced to 50 and 49 ml/kg with L-PRVC601:1 and L-PRVC601:1.5. Compliance was 16 and 18 ml/cmH2O with L-PRVC202:1 and L-PRVC201.5:1, while it was lower with L-IPPV201:1.5, L-PRVC601:1 and L-PRVC601:1.5. Oxygen delivery was maintained at pre-lavage level with L-PRVC201.5:1 (657 ml/min.m2), the other modes displayed reduced oxygen delivery compared with pre-lavage.

CONCLUSION

Neither the rapid frequency modes nor the low frequency volume-controlled mode kept the surfactant deficient lungs open. Pressure-controlled inverse ratio ventilation (20 bpm) kept the lungs open at reduced end-inspiratory airway pressures and hence reduced risk of barotrauma. Reducing I:E ratio in this latter modality from 2:1 to 1.5:1 further improved oxygen delivery.

Alveolar integrity in pulmonary emphysema using technetium-99m-DTPA and technetium-99m-HMPAO radioaerosol inhalation lung scintigraphy

The alveolar integrity (AI) in 17 male patients with pulmonary emphysema (EMPH) diagnosed by chest x-ray was measured by 99mTc-DTPA and 99mTc-HMPAO radioaerosol inhalation lung scintigraphy.

METHODS

The patients were divided into two groups: (A) nine patients with pulmonary emphysema and normal carbon monoxide diffusion capacity (DLCO) and (B) eight patients with pulmonary emphysema and abnormal DLCO. The degree of AI damage in EMPH was presented as the slope of the time-activity curves from the dynamic left lung imagings in DTPA and HMPAO. The AI of EMPH patients were compared with the AI of 16 normal controls.

RESULTS

The results show that: (1) the slope of DTPA is larger than that of HMPAO in each of the portions of the left lung for any of the study groups; (2) statistical differences were found between the normal controls and EMPH patients in HMPAO but not in DTPA; and (3) the correlation was not good between DLCO and DTPA/HMPAO in EMPH patients.

CONCLUSION

Our results suggest that: (1) at least two different mechanisms in the lungs were at work; (2) the AI damage in EMPH developed mainly in the lipophilic part of the alveoli; and (3) the AI damage presented as slopes of DTPA/HMPAO in our study was different from the traditional pulmonary function such as DLCO.

Lung lesions induced by continuous- and pulsed-wave (diagnostic) ultrasound in mice, rabbits, and pigs

These studies documented the presence or absence of macroscopic and microscopic intraparenchymal hemorrhage in individual lung lobes of mice, rabbits, and pigs exposed to continuous- and pulsed-wave (diagnostic) ultrasound; we described the character of and lesions associated with the hemorrhage and compared differences in the lesions among species and exposure conditions to investigate the pathogenic mechanisms and species differences associated with ultrasound-induced lung hemorrhage. In a series of three sequential interdependent studies, 312 mice, 91 rabbits, and 74 pigs were divided at random into experimental groups and exposed to continuous-wave ultrasound (3 kHz modulated at 120 Hz) of acoustic pressure levels ranging from 0 to 490 kPa for 5, 10, or 20 minutes. In a fourth study, three mice, 43 rabbits, and six pigs were divided at random into experimental groups and exposed to pulsed-wave ultrasound (3- and 6-MHz center frequency) of peak rarefactional acoustic pressure levels ranging from 0 to 5.6 MPa for 5 minutes. Macroscopic lesions induced by continuous- and pulsed-wave ultrasound consisted of dark red to black areas of hemorrhage that extended from visceral pleural surfaces into lung parenchyma. Hemorrhage appeared spatially related to the edges of lung lobes where pleura of dorsal and ventral surfaces met, occurred in specific lung lobes in all three species, and appeared anatomically related to lung that was closest to and in contiguous alignment with the ultrasound transducer and thus the path of the sound beam. Macroscopic lesions were similar in all species under all exposure conditions for both continuous- and pulsed-wave ultrasound; however, hemorrhage was not induced in pig lung exposed to pulsed-wave ultrasound at any peak rarefactional acoustic pressure level. Eighteen mice (145 kPa exposure pressure), 60 rabbits (145-460 kPa exposure pressure), and 58 pigs (145-490 kPa exposure pressure) from study 3 were used for microscopic evaluation of lung exposed to continuous-wave ultrasound; three mice (6 MHz; 2.9 and 5.4 MPa), 39 rabbits (3 and 6 MHz; 2.3-5.4 MPa), and six pigs (3 and 6 MHz; 3.3, 5.4, and 5.6 MPa) from study 4 were used for microscopic evaluation of lung exposed to pulsed-wave ultrasound. Microscopic lesions and the character of hemorrhage induced by continuous-wave ultrasound were different from those induced by pulsed-wave ultrasound. Lesions induced by continuous-wave ultrasound under all exposure conditions were similar in all three species. Lesions induced by pulsed-wave ultrasound under all exposure conditions were similar in all three species.(ABSTRACT TRUNCATED AT 400 WORDS)