1088 Thorax 1998;53:1088–1096

Case reports Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from

Spiral CT pulmonary is an Commentary eVective way of demonstrating pulmonary embolism in segmental and larger .4 The basic sign of a filling defect within a well David M Hansell opacified pulmonary is straightforward enough. The case report by Oliver et al2 highlights the fact that there may be ancillary signs of pulmonary embolism on spiral CT The common feature of the reports by Franco scanning—in this case shift of the interven- et al1 and Oliver et al2 is the use of spiral (or volumetric) computed to demon- tricular septum—which corroborates the diag- strate features which would not be readily nosis and, more controversially, provides prog- identifiable on conventional computed tomo- nostic information. Shift of the interventricular graphic (CT) scanning. The advantages of spi- septum and other signs of right ventricular ral CT over conventional CT scanning are dysfunction are readily demonstrated on echo- twofold: increased speed of data acquisition cardiography, but in cases of suspected pulmo- and volumetric (rather than slice by slice) data nary embolism does not acquisition. The attribute of speed means that provide the breadth of information of a spiral most thoracic examinations can be performed CT examination. For example, additional signs within a single breath hold and the timing of of pulmonary embolism, including a mosaic intravenous contrast administration can be perfusion pattern of the lung parenchyma and precisely tailored, thus allowing reproducible radiographically cryptic pleural eVusions or enhancement of any desired part of the small pulmonary infarcts, can be readily picked vasculature—for example, the pulmonary ar- up on spiral CT scanning. Conversely, because teries in cases of suspected pulmonary embo- spiral CT scanning provides the “big picture”, lism. Because an entire volume of data is an alternative diagnosis may be shown by spiral acquired (with almost equal spatial resolution CT scanning in up to one third of patients in the three axes) it is possible to reconstruct investigated for suspected pulmonary images in any plane, including three- embolism.5 dimensional (3-D) reconstructions.3 Most ex- The application of image processing to volu- aminations acquired with spiral CT scanning metric spiral CT data can be broadly divided are presented as a series of transaxial slices, into graphic 3-D realisations—for example, http://thorax.bmj.com/ reflecting the traditional presentation of con- virtual reality bronchoscopy6—and the render- ventional CT images. ing of data so that it is suitable for quantitative In the report by Franco et al1 the clarity with analysis. However, progress towards routine which the anomalous arteries feeding the volumetric (3-D) depictions of spiral CT data sequestrated lung are shown on the 3-D recon- is likely to be slow.7 Even at this early stage of structions is striking. In the past a separate development it is possible to extract very preoperative examination (either aortography precise volumetric measures of abnormal lung; or possibly magnetic resonance angiography) the most obvious application is in the quantifi- to identify the vascular supply would have been cation of low attenuation lung (corresponding on September 25, 2021 by guest. Protected copyright. regarded as mandatory. Other imaging tests to emphysema) on inspiratory and expiratory such as radionuclide or ultra- spiral CT scans. Early results have shown sound may answer specific questions in cases of remarkably good correlation between the pulmonary sequestration, but the wealth of extent of low attenuation lung derived from information now available from a single spiral 3-D reconstructions of the lungs with func- CT examination is remarkable. Quite apart tional indices of air flow obstruction and air from their aesthetic appeal, the main benefit of trapping.8 With this new technique the entire these readily produced 3-D reconstructions is lungs are evaluated, unlike the conventional an easy appreciation of what can be complex “density mask” approach which can be applied anatomy. Nevertheless, claims for the increased only to individual CT sections, (which intro- diagnostic gain from these 3-D reconstructions duces problems with sampling). With the pow- should not be too extravagant: the anomalous erful combination of volumetric data from spi- vessels would be identifiable on images pre- ral CT scanning and advanced image sented in the standard transaxial format, processing, the excitement has only just begun. although without such immediacy. Further- more, demonstration of the venous drainage into the pulmonary circulation (for the classic 1 Franco J, Aliaga R, Domingo ML, et al. Diagnosis of pulmo- nary sequestration by spiral CT angiography. Thorax Royal Brompton intralobar sequestrations) may not be so read- 1998;53:1089–92. Hospital, London ily obtained with a single spiral CT examina- 2 Oliver TB, Reid JH, Murchison JT. Interventricular septal SW3 6NP,UK tion. However, the ability to extract so much shift due to massive pulmonary embolism shown by CT D M Hansell pulmonary angiography: an old sign revisited. Thorax information from a spiral CT examination rep- 1998;53:1092–4. 3 Remy J, Remy-Jardin M, Artaud D, et al. Multiplanar and Correspondence to: resents a substantial advance on conventional three-dimensional reconstruction techniques in CT: im- Dr D M Hansell. CT scanning. pact on chest diseases. Eur Radiol 1998;8:335–51. Commentary 1089

4 Remy-Jardin M, Remy J, Deschildre F, et al. Diagnosis of pul- datasets: a comparison with fibreoptic bronchoscopy. Clin Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from monary embolism with spiral CT: comparison with pulmo- Radiol 1997;52:837–41. nary angiography and scintigraphy. 1996;200:699– 7 Rubin GD, Napel S, Leung AN. Volumetric analysis of 706. volumetric data: achieving a paradigm shift. Radiology 5 Cross JJ, Kemp PM, Walsh CG, et al. A randomized trial of 1996;200:312–7. spiral CT and ventilation perfusion scintigraphy for the 8 Mergo PJ, Williams WF, Gonzalez-Rothi R, et al. diagnosis of pulmonary embolism. Clin Radiol 1998;53: Three-dimensional volumetric assessment of abnormally 177–82. low attenuation of the lung from routine helical CT: 6 Chinn RJS, Mellor J, Yang GZ, et al. Three dimensional inspiratory and expiratory quantification. AJR 1998;170: computed tomography bronchoscopy using clinical 1355–60.

Thorax 1998;53:1089–1092 mal portion of the lung. More recently, other Diagnosis of pulmonary procedures have been advocated as a less inva- sive means of identifying the anomalous artery. sequestration by spiral Although conventional computed tomographic CT angiography (CT) scanning can show both abnormal lung parenchyma and the systemic arterial supply,3 it lacks the multiplanar images which limits its usefulness in the diagnosis of sequestration. José Franco, Roberto Aliaga, María L With the advent of helical technology, spiral Domingo, Pedro Plaza CT angiography is able to delineate the and its branches. The role of CT scanning in evaluating suspected pulmonary sequestration 4 Abstract should therefore be re-evaluated. The diagnosis of pulmonary sequestration In this report we describe the use of spiral traditionally requires arteriography to CT angiography to image the aberrant sys- identify abnormal systemic vessels feed- temic artery in four cases of pulmonary ing the abnormal portion of the lung. sequestration. Non-invasive imaging techniques have recently been used to replace arteriogra- Methods phy. Conventional computed tomographic We performed four CT angiography studies (CT) scanning is, however, at a disadvan- with a PQ2000S helical scanner (Picker Inter- tage because of its inability to obtain national Inc, Highlands Heights, Ohio, USA). multiplanar images. The combination of Spiral volumetric CT scanning was performed slip ring CT scanning and computerised with 4 mm slide thickness, 4 mm table speed, three-dimensional reconstruction (spiral 3 mm reconstruction index, and smooth recon- CT angiography) can be used to visualise struction algorithm. A non-ionic contrast me- the anatomical detail of a wide range of dium (120 ml, iodine 300 mg/ml) was adminis-

vessels within the lung. Four cases of pul- tered at a rate of 3 ml/s via the antecubital . http://thorax.bmj.com/ monary sequestration are reported which were successfully diagnosed using spiral CT angiography. Spiral CT scanning al- lows simultaneous imaging of anomalous vessels and lung parenchyma in a single examination and is particularly useful in the diagnosis and assessment of pulmo- nary sequestration.

(Thorax 1998;53:1089–1092) on September 25, 2021 by guest. Protected copyright.

Department of Keywords: pulmonary sequestration; spiral computed Pneumology tomography J Franco P Plaza Pulmonary sequestration is a rare congenital Department of Radiology pulmonary disorder defined as an area of dys- R Aliaga plastic and non-functioning pulmonary tissue M L Domingo with an anomalous systemic blood supply.1 It has been classically described in two forms— University Hospital intralobar sequestrations located within the Dr. Peset, Valencia, visceral pleura and surrounded by normal Spain lung, and extralobar sequestrations which have Correspondence to: a separate pleural covering. Both types are sup- Dr J Franco, Centro de plied with blood from the aorta or its branches. Especialidades de Alzira, Hort dels Frares 60, The venous return of the intralobar sequestra- E-46600 Alzira, Valencia, tion is usually via the pulmonary while Spain. extralobar sequestrations generally have sys- temic venous drainage. Nevertheless, many Received 23 February 1998 Returned to author variations and combinations of these classical Figure 1 Spiral CT scan of case 1. (A) Contrast 13 May 1998 patterns have been described.2 enhanced CT axial image showing the small aberrant Revised manuscript received Traditionally, the diagnosis of pulmonary artery (arrow) adjacent to the upper abdominal aorta (a). 8 July 1998 (B) Anterior view of helical CT angiogram. The Accepted for publication sequestration requires arteriography to identify anomalous artery (arrow) feeding the pulmonary lesion 20 August 1998 abnormal systemic vessels feeding the abnor- (L) arises from the coeliac axis. 1090 Franco, Aliaga, Domingo, et al Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from

Figure 3 Spiral CT scan of case 3. (A) Contrast enhanced CT axial image showing an anomalous systemic Figure 2 Spiral CT scan of case 2. (A) Contrast enhanced artery (arrow) feeding the pulmonary lesion (L) in the left CT axial image showing the pulmonary mass (M) supplied lower lobe. (B) Left oblique view of helical CT angiogram by an anomalous systemic artery (arrow) arising from the demonstrating the anomalous artery (arrow) originating upper abdominal aorta (a). (B) Right view of helical CT from the left wall of the thoracic aorta (a). angiogram showing the course of the arterial vessel (arrow) to the pulmonary mass (M). Note the venous drainage (arrowhead) to the pulmonary veins (V). ous to the diaphragm and identified its anoma- lous arterial supply derived from the upper A 25 second scan delay was used in order to abdominal aorta just above the coeliac artery; optimise contrast in the systemic arterial phase venous drainage into the pulmonary veins was

of the study. Three-dimensional reconstruction also visualised. Intralobar pulmonary seques- http://thorax.bmj.com/ (3D) was performed with a Voxel Q work station tration was confirmed by aortography and tho- using a shaded surface display (SSD) program racic surgery. with segmentation option. CASE 3 Case reports A 65 year old male cigarette smoker had a two CASE 1 week history of productive cough and fever. A 32 year old man with a 34 pack-year history The patient improved with antibiotic therapy of cigarette smoking who still smoked two but a persistent cough developed. packs a day was admitted to the smoking of the chest showed partial collapse of the left on September 25, 2021 by guest. Protected copyright. cessation programme at our hospital. There lower lobe and a mass like opacity with an air- was a history of pneumonia at the age of 14. A fluid level. A CT scan demonstrated a showed localised air trapping non-homogeneous mass with multiple cystic in the left lower lobe. A spiral CT scan (fig 1) appearing spaces and cavitation involving the revealed a multicystic lesion in the posterior posterior basal segment of the left lower lobe. basal segment of the left lower lobe supplied by Fibreoptic bronchoscopic examination dis- an artery derived from the coeliac axis; venous closed no abnormality. A percutaneous fine return to the pulmonary veins was also demon- needle aspiration biopsy specimen of the lesion strated. The presence of the anomalous revealed non-diagnostic findings. Spiral CT systemic artery and venous drainage were con- angiography (fig 3) showed a feeding systemic firmed by aortography. The patient was artery arising from the descending thoracic asymptomatic and refused surgery. aorta and venous drainage to the pulmonary veins. At surgery an infected intralobar seques- CASE 2 tration was found. A 28 year old man presented with a one month history of recurrent haemoptysis. He smoked CASE 4 one pack of cigarettes daily. Chest radiography A 41 year old man was admitted to hospital showed a mass, 4 cm in diameter, in the right with a 48 hour history of fever and pleuritic lower lobe. Bronchoscopic examination indi- chest pain. There was no history of use of cated that the source of bleeding was the right tobacco. Chest radiography revealed a homo- lower lobe but no endobronchial lesions were geneous density in the right lower lobe. He was seen. A spiral CT scan (fig 2) revealed a homo- diagnosed as having pneumonia and treated geneous mass in the right lower lobe contigu- with clarithromycin. One month later he was Diagnosis of pulmonary sequestration by spiral CT angiography 1091

questrations are frequently discovered during Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from the neonatal period in infants with other congenital anomalies. The clinical picture is usually dominated by the associated anomalies although infection can occur, especially if there is a communication with the oesophagus or the stomach. Extralobar sequestrations that are not diagnosed in newborn infants are often asymp- tomatic and detected on routine radiography. In case 1 pulmonary sequestration was detected by routine radiography, while in case 2 haemoptysis was the presenting symptom of an intralobar sequestration. Case 3 had an intralobar sequestration typically manifested by symptoms of infection and case 4 had con- comitant disseminated tuberculosis. Plain radiographs of the chest often show a single homogeneous opacity or, less com- monly, a cystic mass in the base of one lung that can sometimes suggest the diagnosis of sequestration.1 Less specific findings include recurrent pneumonia and focal bronchiectatic changes. The principal objective for diagnosis of pulmonary sequestration is to identify the systemic artery supply. With this information, imaging can distinguish sequestration from other causes of lung opacity. Because accessory Figure 4 Spiral CT scan of case 4. (A) Contrast enhanced CT axial image showing a systemic artery arteries, pleural investment, and venous drain- (arrowhead) feeding the pulmonary lesion (L) in the right age are adequately determined intraopera- lower lobe. (B) Right oblique view of helical CT angiogram tively, at some institutions only the presence showing the anomalous systemic artery (arrow) arising from the coeliac axis. a = aorta. and location of an aberrant systemic artery are considered essential for preoperative assess- readmitted with clinical deterioration, unpro- ment for any symptomatic pulmonary ductive cough, weakness, anorexia, weight loss, sequestration.4 and fever. Chest radiography showed a diVuse Imaging strategies for suspected pulmonary and bilateral micronodular pattern with per- sequestration are based on case reports or sistence of the right lower lobe density. Miliary small series since it is a rare congenital disorder tuberculosis was confirmed by histopathologi- and no study exists that objectively compares http://thorax.bmj.com/ cal examination of a transbronchial biopsy imaging techniques for detection, definition, or specimen and positive culture of bronchial cost eVectiveness.4 Since the definitive step in aspirate. A spiral CT angiogram (fig 4) the diagnosis of sequestration is the demon- revealed a focal area of increased density in the stration of the systemic arterial supply, for a right lower lobe supplied by an artery originat- long time diagnosis was made by conventional ing from the coeliac axis. An abdominal aorto- angiography. More recently all imaging tech- gram showed the anomalous artery and the niques capable of showing the artery have been venous return via the pulmonary veins. Be- implicated in evaluating sequestration. Mag- cause the lower pulmonary veins were not netic resonance (MR) imaging and MR angio- on September 25, 2021 by guest. Protected copyright. included in the upper sections of pre- graphy can be used together to diagnose established contrast enhanced helical CT pulmonary sequestration in a single non- acquisition, in this case spiral CT angiography invasive examination.5 Nevertheless, MR can- could not identify the venous drainage. Surgery not accurately evaluate lung parenchyma and was not considered at the time this case was the airways and must be considered in terms of presented. cost and availability. Sonography requires a favourable acoustic window and is ideally Discussion suited for evaluating the chest prenatally and Both intralobar and extralobar sequestration postnatally.6 Other non-invasive techniques for characteristically involves the lower lobes of the evaluation of sequestration such as scintigra- lungs. Intralobar pulmonary sequestration ac- phy are only rarely necessary. counts for 73% of all sequestrations and has a In all four cases described in this report spi- predilection for the posterior basal segment of ral CT angiography successfully delineated the the lower lobes. It occurs slightly more often in origin and course of the anomalous systemic the left lung than in the right.2 The blind end- artery. Axial images were enough to make the ing bronchi, which may become distended diagnosis but three-dimensional reconstruc- trapping mucus, are prone to infection. In tion aided both radiologists and referring clini- about two thirds of the cases reported the first cians by demonstrating anatomical relation- symptoms occur after the age of 10 years and ships, particularly for vessels orientated in the z are usually secondary to a superimposed infec- axis.7 On the other hand, venous drainage was tion. Productive cough, fever, haemoptysis, also identified in the three cases in which lower recurrent pneumonia, and chest pain are pulmonary veins were included in contrast typical presenting complaints. Extralobar se- enhanced helical CT scans. We have performed 1092 Franco, Aliaga, Domingo, et al

three-dimensional reconstruction (3D-SSD) the patient to ionising radiation and the Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from segmentation for a better understanding of the administration of intravenous contrast mate- anatomy of the abnormal systemic arteries. rial. However, as shown in case 2, three- In summary, we report four cases of pulmo- dimensional imaging dedicated to the venous nary sequestration successfully diagnosed drainage can also be made.8 using spiral CT angiography. By allowing Spiral CT angiography is a minimally simultaneous imaging of anomalous vessels invasive technique for vascular imaging that is and parenchymal lesions in a single examina- made possible by combining slip ring CT tion, spiral CT angiography is a particularly scanning and computerised three-dimensional eYcacious technique and has the potential to reconstruction.9 Spiral CT angiography has become the procedure of choice in the diagno- several advantages over other non-invasive vas- sis and assessment of pulmonary sequestration. cular imaging techniques.410 CT scanning, with its superior spatial resolution, yields the 1 Felker RE, Tonkin ILD. Imaging of pulmonary sequestra- most information about the bronchial anatomy tion. AJR 1990;154:241–9. and the pulmonary parenchymal lesion. 2 Savic B, Birtel FJ, Tholen W, et al. Lung sequestration: report of seven cases and review of 540 published cases. Sonography and MRI cannot evaluate lung Thorax 1979;34:96–101. abnormalities accurately although MRI can 3 Ikezoe J, Muruyama S, Godwin JD, et al. Bronchopulmo- nary sequestration: CT assessment. Radiology 1990;176: reveal the cystic nature of many intralobar 375–9. sequestrations as well as the variable solid, 4 Frush DP, Donnelly LF. Pulmonary sequestration spectrum: a new spin with helical CT. AJR 1997;169:679– fluid, haemorrhagic, and mucus-containing 82. components. MR angiography is hampered by 5 Doyle AJ. Demonstration of blood supply to pulmonary sequestration by MR angiography. AJR 1992;158:989–90. artefacts caused by respiratory motion whereas 6 May DA, Barth RA, Yeager S, et al. Perinatal and postnatal this problem is generally avoided in helical CT chest sonography. Radiol Clin North Am 1993;31:499–516. 7 Johnson PT, Fischman EK, Duckwall JR, et al. Interactive scanning. CT angiography is less expensive three-dimensional volume rendering of spiral CT data: than MR angiography and can be used on current applications in the thorax. Radiographics 1998;18: 165–87. patients with a metallic device or who do not 8 Rémy J, Rémy-Jardin M. Spiral CT angiography of pulmo- tolerate the MR examination. Furthermore, nary vessels. In: Rémy-Jardin M, Rémy J, eds. Spiral CT of the chest. Berlin: Springer-Verlag, 1996: 231–64. helical CT scanning is faster resulting in less 9 Dillon EH, van Leeuwen MS, Fernandez MA, et al. Spiral sedation and reduced amount of contrast CT angiography. AJR 1993;160:1273–8. 10 Amitai M, Konen E, Rozenman J, et al. Preoperative evalu- medium. The disadvantages of helical CT ation of pulmonary sequestration by helical CT angio- scanning are minor and arise from exposure of graphy. AJR 1996;167:1069–70.

Thorax 1998;53:1092–1094 Keywords: pulmonary embolism; right ventricular dys- Interventricular septal function; spiral computed tomography; pulmonary angiography shift due to massive http://thorax.bmj.com/

pulmonary embolism A 43 year old man collapsed while out walking. On admission to hospital he was dyspnoeic and shown by CT cyanosed. On direct questioning he admitted to right leg pain. Examination showed that his Department of Clinical pulmonary angiography: heart rate was 105 beats/min, respiratory rate Radiology, University was 28 breaths/min, and blood pressure was Hospital of Wales, an old sign revisited 100/60 mm Hg. His jugular venous pressure

CardiV,UK on September 25, 2021 by guest. Protected copyright. T B Oliver was raised but examination was otherwise nor- mal. Electrocardiography demonstrated sinus T B Oliver, J H Reid, J T Murchison Department of tachycardia; the chest radiograph was normal. Radiology, Borders Measurement of arterial blood gas tensions General Hospital, confirmed hypoxaemia with hypocapnia (PO Melrose, UK 2 CO J H Reid Abstract 8 kPa on 6 l/min oxygen, P 2 4 kPa). An The computed tomographic (CT) pulmo- echocardiogram demonstrated dilatation of the Department of Clinical nary angiogram appearances of acute right ventricle. The clinical features of syncope, Imaging, Royal right ventricular dysfunction due to mas- cyanosis and dyspnoea with engorged neck Infirmary of sive pulmonary embolus in a patient are veins in a patient with a normal chest Edinburgh, radiograph and clinical suspicion of deep Edinburgh, UK described. Abnormal findings comprised J T Murchison right ventricular dilatation, interven- venous thrombosis led to a presumptive tricular septal shift, and compression of diagnosis of pulmonary embolus. Correspondence to: the left ventricle. These changes resolved A computed tomographic (CT) pulmonary Dr JT Murchison, angiogram was performed. A 3 mm spiral scan, Department of Clinical following thrombolysis. Use of CT pulmo- Imaging, Royal Infirmary of nary angiography to diagnose pulmonary reconstructed at 1.5 mm intervals, was under- Edinburgh, Lauriston Place, emboli is increasing. Secondary cardiac taken on a Hi Speed Advantage scanner (Gen- Edinburgh. EH3 9YW, UK. eVects are established diagnostic features eral Electric Medical Systems, Milwaukee, Wisconsin, USA) using 150 ml of contrast Received 5 March 1998 shown by echocardiography. These have Returned to author not been previously described but are (200 mg I/ml) at 4 ml/s. This showed multiple 13 May 1998 important to recognise as they may carry pulmonary emboli within the main and seg- Revised manuscript received mental pulmonary arteries. In addition there 3 July 1998 important prognostic and therapeutic im- Accepted for publication plications. was dilatation of the right ventricle and atrium 20 August 1998 (Thorax 1998;53:1092–1094) with normal wall thicknesses, the interven- Interventricular septal shift due to pulmonary embolism on CT angiogram 1093

initial abrupt rise in pulmonary artery pressure Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from causes increased right ventricular afterload which results in right ventricular dilatation and dyskinesia. Secondary eVects of this are tricus- pid regurgitation, right atrial enlargement, and loss of respiratory variation in calibre of the great veins. Increased right ventricular wall tension may reduce local coronary blood flow, resulting in ischaemia which impairs right ven- tricular function further. As the right ventricle dilates, the interventricular septum is displaced towards the left ventricle, the right ventricle assuming a circular axial configuration and the left ventricle a crescentic appearance more typical of the normal right ventricle. This sep- tal shift, combined with the constraining influ- ence of the pericardium, results in reduced left ventricular filling which is already compro- mised by reduced preload. The cardiac output falls. Signs documenting this sequence such as right ventricular dilatation and hypokinesis (which may spare the apex), abnormal inter- ventricular septal motion, pulmonary artery dilatation, tricuspid regurgitation, and loss in respiratory variation in inferior vena caval diameter can be detected at echocardiography. Echocardiographic assessment of the right ventricle has been recommended as an integral part of the investigative algorithm for suspected acute pulmonary embolus published recently Figure 1 Computed tomographic (CT) pulmonary 2 angiograms at the level of the interventricular septum in a by the British Thoracic Society working party. patient with massive pulmonary embolism. (A) The scan at “Right ventricular dysfunction” is an umbrella presentation shows right ventricular dilatation and septal term which also includes more subjective displacement which results in compression of the left ventricle. A filling defect due to an embolus is seen within a echocardiographic findings such as abnormali- segmental branch of the descending right pulmonary artery ties of the motion of the right ventricular wall. (arrow). (B) Five days after thrombolysis the pulmonary These findings are often encountered in lesser

embolus has resolved. The associated reduction in http://thorax.bmj.com/ pulmonary artery pressure has allowed normal septal and degrees of pulmonary embolus and have led to ventricular appearances to return. The long axis of the debate over the significance of the more objec- heart now occupies a more normal position. RV = right tive signs. Acute right ventricular dilatation and ventricle, LV = left ventricle, S = interventricular septum. interventricular septal shift have been associ- tricular septum was displaced to the left, and ated specifically with massive pulmonary there was compression of the left ventricle (fig embolism34and reversible septal displacement 1A). These features persisted throughout the has been described in a series of patients cardiac cycle. A central venous was requiring aggressive treatment for circulatory placed and tissue plasminogen activator was failure due to massive pulmonary embolism.3 infused into the central pulmonary arteries. Thrombolysis is an accepted treatment in mas- on September 25, 2021 by guest. Protected copyright. Immediately before treatment the central sive life threatening pulmonary embolus and its venous pressure was 22 mm Hg, right atrial administration in the case described here was pressure was 30 mm Hg, right ventricular associated with a rapid return of pulmonary pressure was 33/13 mm Hg, and pulmonary and systemic arterial pressures towards nor- artery pressure was 33/20 mm Hg. Five hours mal. Recognition of the signs presented by CT after treatment the pulmonary artery pressure scanning or echocardiography allows more had reduced to 20/10 mm Hg and systemic aggressive therapy to be targeted to individuals blood pressure had increased to 160/70 mm at greatest risk. Hg. A continuing anticoagulation regime was CT pulmonary angiography is increasingly commenced. A venogram showed thrombus used to diagnose pulmonary embolus. It is within the right popliteal vein. A repeat CT non-invasive and quick to perform. In the case angiogram five days after treatment showed described the patient was imaged directly after considerable reduction in the load of embolic initial assessment in the emergency room and material within the pulmonary arteries to- spent less than 15 minutes in the imaging suite. gether with a return of the interventricular sep- Comparative studies have shown excellent cor- tum to its normal position and resumption of relation between CT and conventional pulmo- normal right and left ventricular morphologies nary angiography in the detection of emboli in (fig 1B). segmental or larger vessels and in many centres the technique has largely replaced conventional Discussion pulmonary angiography.56 Secondary signs of Massive pulmonary embolism sets in sequence pulmonary embolus have not, to our knowl- a chain of physiological events that ultimately edge, been described at CT pulmonary angio- lead to reduced systemic cardiac output.1 The graphy. The interventricular septum is usually 1094 Letters, Book reviews, Notices, Correction

clearly visualised by thoracic CT scanning fol- throughout the CT scan is likely to be a reliable Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from lowing intravenous contrast. We have observed indication that an embolus of major propor- interventricular septal shift in several patients tions has occurred. with acute pulmonary embolus. Septal shift may also be identified by CT scanning or MRI 1 Goldhaber SZ. Pulmonary embolism. In: Braunwald E, ed. in patients with chronic pulmonary hyper- Heart disease:a textbook of cardiovascular medicine. Vol 2. 2nd tension due to a variety of causes; however, an ed. Philadelphia: WB Saunders, 1997: 1582–603. important distinguishing feature in such cases 2 British Thoracic Society, Standards of Care Committee. Suspected acute pulmonary embolism: a practical ap- is co-existing thickening of the right ventricular proach. Thorax 1997;52(Suppl 4). wall, which is not observed in acute pulmonary 3 Jardin F, Dubourg O, Gueret P, et al. Quantitative two-dimensional echocardiography in massive pulmonary embolus and was not apparent in the case pre- embolism: emphasis on ventricular interdependence and sented here. A typical CT pulmonary angio- leftward septal displacement. J Am Coll Cardiol 1987;10: 1201–6. gram will include in its acquisition time two or 4 Kasper W, Geibel A, Tiede N, et al. Distinguishing between three cardiac cycles and some normal variation acute and subacute massive pulmonary embolism by in the appearance of the cardiac chambers is to conventional and . Br Heart J 1993;70:352–6. be anticipated over the length of the scan. 5 Blum AG, Delfau F, Grignon B, et al. Spiral computed tom- Nevertheless, the constellation of CT findings ography versus pulmonary angiography in the diagnosis of acute massive pulmonary embolism. Am J Cardiol 1994;74: of proximal emboli, enlargement of the right 96–8. ventricle with normal wall thickness, interven- 6 Remy-Jardin M, Remy J, Deschildre F, et al. Diagnosis of pulmonary embolism with spiral CT: comparison with tricular septal shift, and crescentic axial left pulmonary angiography and scintigraphy. Radiology 1996; ventricular morphology which persists 200:699–706.

example, a prospective randomised control- BERTIL LINDMARK led trial, and the results must be interpreted Department of Clinical Drug LETTERS TO with great caution. Safety and Epidemiology, Thirdly, no support for an association Clinical R&D, between bambuterol and an increased risk for Astra Draco, THE EDITOR S221 00 Lund, cardiac failure has been found in our review Sweden of preclinical studies, clinical studies (includ- ing >3000 patients/healthy volunteers), or post-marketing surveillance (based on >130 1 Martin RM, Dunn NR, Freemantle SN, et al. Cardiac risks with â million treatment days). The risk of non-fatal cardiac failure and Fourthly, according to the authors there ischemic heart disease with long acting â2 ago- have been no spontaneous reports of cardiac nists. Thorax 1998;53:558–62.

agonists http://thorax.bmj.com/ failure with bambuterol to the Committee on Martin et al1 suggest that caution should be Safety of Medicines. This is in agreement AUTHORS’ REPLY Bertil Lindmark of Astra Draco makes five points about our study on exercised when prescribing long acting oral â with the WHO database Intdis, with no agonists to patients at risk of cardiac failure, reports of cardiac failure for bambuterol. the risk of non-fatal cardiac failure and ischaemic heart disease with long acting based on results from a prescription event Finally, the paper suggests a doubled â2 agonists. Firstly, he points out that all ago- monitoring (PEM) study. asthma mortality in patients receiving salm- â2 nists should be used with caution in patients Firstly, all â2 agonists (short and long eterol. In our opinion the reported higher acting, oral and inhaled) should be used with relative risk for non-fatal cardiac failure for with severe cardiovascular disease. The car- 1 caution in patients with severe cardiovascular bambuterol and the doubled asthma mor- diac eVects of â2 agonists are well described, disease, as is pointed out in the package insert tality for salmeterol both appear equally but there is limited evidence available on for all drugs of this class. explicable by factors other than direct causal- whether or not the risks of adverse cardiac on September 25, 2021 by guest. Protected copyright. Secondly, the study does not provide any ity, such as confounding by concomitant dis- eVects diVer depending on the dose and evidence on this issue. PEM studies are not eases and disease severity. method of administration of the drug. designed to study causal relations but to gen- Thus, PEM data may be of help in identi- Clearly, these are important questions for erate new hypotheses. Although the authors fying signals with new drugs, but there is lit- prescribing doctors faced with treating asth- have made an attempt to consider several tle if any merit in comparing drugs used in matic patients with concomitant cardiac potential biases in the analyses, the study diVerent populations and introduced to the disease. An observational cohort study design is inappropriate compared with, for market at diVerent times. formed from health insurance databases from the Province of Saskatchewan, Canada found Table 1 Rates of cardiac failure during the first month of exposure to bambuterol or a cardiovascular an increased relative risk of death from drug studied by prescription event monitoring (PEM) cardiovascular disease in users of â2 agonists taken orally or by nebuliser, but not in users

No. of of â2 agonists administered by metered dose No. of patients with patient-months Rate (events per inhaler.2 The deaths occurred in patients with Drug* (ranked reported cardiac failure of exposure 1000 months of Mean (SD) significant cardiac disease, suggesting that by rate) during month 1 during month 1 exposure) age Males (%) â2 agonists taken orally or by nebuliser should Xamoterol 97 4 463 21.7 70.8 (13.9) 53.0 be avoided in patients at high risk of Nicorandil 73 11 578 6.3 66.9 (11.2) 61.1 cardiovascular events. We found that the oral Bambuterol 29 5 891 4.9 58.5 (18.6) 44.8 â agonist bambuterol, but not the inhaled â Losartan 53 12 990 4.1 63.5 (12.1) 40.2 2 2 Diltiazem 24 8 808 2.7 62.3 (13.9) 59.3 agonist salmeterol, was associated with an Enalapril 33 13 544 2.4 61.2 (14.9) 46.1 increased risk of non-fatal cardiac failure. Perindopril 17 8 368 2.0 61.8 (12.7) 45.0 The results from both these studies are plau- Nicardipine 17 9 517 1.8 62.9 (13.9) 48.4 sible as oral â agonists provide a greater sys- Lisinopril 18 11 574 1.6 60.9 (14.3) 44.0 temic dose than that achieved with metered Ramipril 2 1 277 1.6 60.5 (12.4) 45.1 dose inhalers1 and tachycardia and prolonged Amlodipine 12 12 085 1.0 61.8 (14.7) 46.9 Q-T interval have been reported principally *Betaxolol, doxazosin, isradipine not shown as number of patients with event was <2, or rate was <1.0 per with nebulised or oral â agonists.2 The 1000 patient-months of exposure. advised total daily dose of oral bambuterol Letters, Book reviews, Notices, Correction 1095

(20 mg) is 200 times that of inhaled salm- 6 Li Wan Po A, Zhang WY. What lessons can be logical tests performed give important infor- Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from eterol (100 µg).3 learnt from withdrawal of mibefradil from the mation on the prevalence of infection, but are market? Lancet 1998;351:1829–30. Secondly, Dr Lindmark reiterates our point 7 Lawson DH. Pharmacovigilance in the 1990s. not suYciently complete to make definitive that the results must be interpreted with Br J Clin Pharmacol 1997;44:109–10. conclusions on the incidence of acute C caution because the study was observational, 8 Martin RM, Kapoor KV, et al. Underreporting pneumoniae infection in the populations of adverse drug reactions to newly marketed, and more definitive evidence would come black triangle drugs in general practice in under study. The major pitfall in the study, as from a prospective randomised trial. England: observational study. BMJ 1998;317: pointed out by the authors, is the small Nevertheless, hypotheses about drug safety 119–20. proportion of patients from whom a convales- concerns are often generated from observa- cent serum sample was drawn. Moreover, the tional studies.4 Such studies drive further arbitrary exclusion of IgM positive patients research because they provide an “a priori” Genetics and tuberculosis for the diagnosis of acute C pneumoniae infec- hypothesis and allow the formulation of a tion may have been misleading since the clinically relevant end point. Until results from Dr Richard Bellamy alludes to the important possibility of cross reactivity with rheumatoid prospective trials become available, observa- fact, frequently ignored by immunologists, factor could have been eVectively ruled out tional research using cohort or case-control geneticists and epidemiologists, that tubercu- by using IgG absorption prior to IgM micro- 1 techniques remains an important source of losis has several diVerent clinical forms. Phy- immunofluorescence determination.2 Not- evidence about the safety of drugs. sicians have emphasised the diVerence be- withstanding these facts, the authors con- Thirdly, he states that a review by Astra tween primary tuberculosis, which is clude that the study does not support “an Draco has found no evidence from pre- comparable to Lurie’s susceptible rabbits association between C pneumoniae antibody marketing or post-marketing studies of an with disseminated disease, and post primary titres and the incidence of acute asthma association between bambuterol and cardiac tuberculosis, best characterised by smear attacks”. failure. In general, pre-marketing studies positive pulmonary tuberculosis and Lurie’s Analysis of table 1 indicates that the acute have their own limitations,5 as evidenced by “resistant” rabbits. HLA associations with asthma and control populations appear to be the recent withdrawal on safety grounds of tuberculosis have indeed been inconsistent significantly diVerent in terms of age and sex two newly launched drugs.6 Similarly, diVer- when all forms of tuberculosis are included. distribution, the control population being ent types of post-marketing surveillance However, the HLA association with DR2, significantly older and showing a male studies, including PEM, have diVerent ad- and particularly with its subtype DR15 in predominance. Both these factors are associ- vantages and disadvantages and, in general, linkage disequilibrium with DQ5, was found ated with increased C pneumoniae incidence only in patients with smear positive pulmo- and prevalence. The authors report using a one system cannot be relied upon to provide 23 all the evidence needed.7 This point also nary tuberculosis. These observations have logistic regression modelling method in applies in response to the fourth comment. In been refined using DNA based HLA typing which the age value is implemented as “± 10 and have confirmed a link with the genes years”, which is roughly equivalent to the dif- particular, it should be noted that there is 4 gross under-reporting of suspected adverse DRB1*1501 and DQB1*0502. Antibody lev- ference in mean age between the acute drug reactions to the Committee on Safety of els to epitopes of the 38kDa antigen of Myco- asthma and control populations. Medicines8 and other regulatory authorities, bacterium tuberculosis restricted antigens were This study is certainly noteworthy in that it higher, suggesting an enhanced immune underlines an association between C pneumo- and there are many diYculties associated 3 with interpreting data from spontaneous responsiveness in those with HLA-DR15. niae infection and severe chronic asthma, reporting schemes.5 The relative importance of the genes involved particularly “brittle” asthma, which will Finally, as is stated clearly in our paper, it is in susceptibility can be assessed by the gene require further investigation in the future. frequency, but also by the attributable risk— possible that the association may be ex- FRANCESCO BLASI plained by factors such as confounding by that is, how much of the disease can be LUIGI ALLEGRA concomitant disease and disease severity. attributed to the presence or absence of a PAOLO TARSIA Interestingly, the rate of cardiac failure particular gene (34% with 95% confidence Istituto di Tisiologia e Malattie dell’Apparato

intervals of 16 to 43% were suggested for Respiratorio, http://thorax.bmj.com/ associated with bambuterol in the first month 3 of treatment was higher than for 11 cardio- DR15 in one population ). Università degli Studi di Milano, Pad. Litta, vascular drugs previously studied by PEM The Lurie experiment suggests that a com- parison between patients with diVerent forms IRCCS Ospedale Maggiore di Milano, (table 1). Only two cardiac drugs, including Italy the inotropic sympathomimetic xamoterol of tuberculosis, matched by ethnic origin, (licensed for use in mild heart failure) had may be valuable in identifying candidate genes for susceptibility to tuberculosis. Since higher rates of cardiac failure. Since it is 1 Cook PJ, Davies P, TunnicliVeW,et al. Chlamy- highly unlikely that the rate of cardiovascular smear positive pulmonary tuberculosis is dia pneumoniae and asthma. Thorax disease in the bambuterol cohort was higher responsible for transmission of the disease, an 1998;53:254–9. 2 Verkooyen RP, Hazemberg MA, Van Haaren than in cohorts of patients taking cardiac understanding of its pathogenesis will be especially important in finding new ways to GH, et al. Age-related interference with Chlamy- drugs, and the bambuterol cohort was the dia pneumoniae microimmunofluorescence se- on September 25, 2021 by guest. Protected copyright. youngest, these data provide further evidence control tuberculosis. rology due to circulating rheumatoid factor. J Clin Microbiol 1992;30:1287–90. that an association cannot be discounted. GRAHAM H BOTHAMLEY Our findings require confirmation, but we East London Tuberculosis Service, remain concerned about the size and biologi- Homerton Hospital, cal plausibility of the association. London E9 6SR, UK Chlamydia pneumoniae RICHARD M MARTIN 1 Bellamy R. Genetic susceptibility to tuberculosis NICHOLAS R DUNN in human populations. Thorax 1998;53:588– and asthma RONALD D MANN 93. SHAYNE N FREEMANTLE 2 Brahmajothi V, Pitchappan RM, Kakkanaiah I read with interest the recent report by Cook VN, et al. Association of pulmonary tuberculo- Drug Safety Research Unit, et al1 in which they report that, compared Bursledon Hall, sis and HLA in South India. Tubercle 1991;72: Blundell Lane, 123–32. with hospital controls, outpatients with 3 Bothamley GH, Schreuder GMT. Human chronic severe asthma had significantly more Southampton SO31 1AA, leukocyte antigen, tuberculosis and Mycobacte- UK rium tuberculosis-specific antibody. J Infect Dis C pneumoniae antibody titres (IgG 64–256 1992;165:598. and/or IgA >8) indicating previous infection, 4 Meyer CG, May J, Stark K. Human leukocyte whereas unselected patients admitted to hos- 1 Nelson HS. â-adrenergic bronchodilators. N antigens in tuberculosis and leprosy. Trends pital for acute asthma attacks had titres simi- Microbiol 1998; :148–54. Engl J Med 1995;333:499–506. 6 lar to controls. They also found that serologi- 2 Suissa S, Hemmelgarn B, Blais L, et al. Bronchodilators and acute cardiac death. Am J cal evidence of acute (re)infection (presence Respir Crit Care Med 1996;154:1598–602. Chlamydia pneumoniae of IgM, a fourfold change in titre, and/or IgG 3 Joint Formulary Committee. British National titre >1:512) was equal among groups. Formulary Number 33. London: British Medical These data are in accord with previous evi- Association and Royal Pharmaceutical Society and asthma of Great Britain, 1997:125–6. dence suggesting an important role for 4 Olsen JH. Interpretation in drug epidemiology. The paper by Cook et al1 examines the possi- chronic C pneumoniae infection as a promoter Lancet 1998;352:162–3. ble association between Chlamydia pneumo- of asthma symptoms but a lesser role for 5 Waller PC, Coulson RA, Wood SM. Regulatory niae infection and asthma. The authors acute infection as a cause for asthma pharmacovigilance in the United Kingdom: 2 current principles and practice. Pharmacoepide- conclude that their data do not support this exacerbations. An additional recent report of miol Drug Safety 1996;5:363–75. association. However, we feel that the sero- positive therapeutic responses to antibiotics 1096 Letters, Book reviews, Notices, Correction

in severe steroid dependent asthmatic pa- In the patient we reported the association (Pp 942; hardback; $150.00). London: Thorax: first published as 10.1136/thx.53.12.1094 on 1 December 1998. Downloaded from tients (aged 13–65) further supports the was with cryptogenic fibrosing alveolitis Academic Press, 1998. 0 12 079027 0. possibility that antibody titres indicative of (CFA) alone whereas, interestingly, the three “previous infection” may also indicate per- patients they report had CFA associated with This is the third edition of an established sistent chronic infection.3 other systemic autoimmune disorders. The book. Aiming to bring together all the recent Acute primary (presence of IgM) or second- fact that CFA alone may be associated with B information on basic mechanisms of asthma ary (fourfold change in titre without IgM) cell lymphomas, and the poorer prognosis and also cover clinical aspects and therapy in C pneumoniae infection has been reported to seen by Nicholson and Corrin in their depth, this is achieved successfully. The scope initiate asthma in previously non-asthmatic patients, as well as ours, supports the hypoth- of the book provides accessible reviews of all individuals.4 Since the incidence of asthma in esis that chronic local stimulation of the lym- facets of asthma, from epidemiology and adults is very small (around one per 1000 per phoid system may play an important part in physiology to allergen avoidance, including year) it is likely that most of the acute exacer- the aetiology and prognosis of these tumours. recent developments in these fields. Modifi- bations occurred in patients who had had pre- T R ORCHARD cations to the popular second edition include vious wheezing episodes. It would be interest- C D ERAUT separate chapters on mediator antagonists ing to know whether Cook et al can AGDAVISON and immunomodulators with consideration retrospectively identify any patients who had Southend Hospital, of the potential therapeutic benefits of inter- their very first wheezing episode; this might be WestcliV on Sea, vening in the complex inflammatory and easier in general practice than in a hospital Essex SS0 0RY, pharmacological pathways systematically based study. UK covered in previous chapters. A new chapter DAVID L HAHN on the pharmacoeconomics of asthma treat- Arcand Park Clinic, ments provides a pertinent reminder that, Dean Medical Center, after the wonders of basic science and the 3434 E Washington Avenue, development of beneficial interventions, a Madison, BOOK REVIEWS wider perspective is required to successfully WI 53704, USA deliver benefits to those who require them. The addition of colour plates provides a wel- 1 Cook PJ, Davies P, TunnicliVeW,et al. Chlamy- come change to the previous black and white prints of the old edition which look a little dia pneumoniae and asthma. Thorax Göran Heden- 1998;53:254–9. Respiratory Measurement. drab in retrospect. 2 Hahn DL. Intracellular pathogens and their role stierna. (Pp 184, paperback; £19.95 (UK), Well written by authorities in their fields in asthma: Chlamydia pneumoniae in adult £22.00 (overseas)). London: BMJ Books, patients. Eur Respir Rev 1996;6:224–30. and uniformly edited with an attractive pres- 3 Hahn D, Bukstein D, Luskin A, et al. Evidence 1998. ISBN 0 7279 1207 0. entation, this is an excellent book which suc- for Chlamydia pneumoniae infection in steroid- ceeds in linking the rapidly developing body dependent asthma. Ann Allergy Asthma Immu- A large amount of information has been of knowledge on asthma with current treat- nol 1998;80:45–9. packed into the 184 pages of this new guide- 4 Hahn D. Incident wheezing and prevalent ment, while keeping the future constantly in asthma have diVerent serologic patterns of book in the Principles and Practice Series. mind.—AF “acute” Chlamydia pneumoniae antibodies in This is a comprehensive review of the princi- adults. Proceedings of the Third Meeting of the ples of ventilation and gas exchange with spe- European Society for Chlamydia Research, Vi- enna,Austria. Bologna: Società Editrice Escula- cial emphasis on the application of pulmo- pio, 1996: 226. nary function measurement during anaesthesia. The book details physiological principles and gives practical measurement CORRECTION guidance, with common sources of error, in

the normal circumstances and during anaes- http://thorax.bmj.com/ Non-Hodgkin’s lymphoma thesia. The content is concise, the style direct with CFA and occasionally hard going. The text is clear and the diagrams are worth a special mention Clinical features of non-smokers We read with interest the case report by for their clarity and simplicity. This is not a with á -antitrypsin deficiency Orchard et al on non-Hodgkin’s lymphoma textbook for beginners and requires a moder- 1 arising in cryptogenic fibrosing alveolitis ate familiarity with the principles of respira- The authors of the paper entitled “Clinical (CFA).1 Although the authors state that this tory physiology, and the rules which govern features and prognosis of life time non- has not been described previously, we re- respiratory mechanics and gas measurement. smokers with severe á1-antitrypsin defi- cently reported six cases of pulmonary B cell This guide represents excellent value for ciency” by N Seersholm and A Kok-Jensen, non-Hodgkin’s lymphomas arising in patients money and would be equally at home in the which appeared on pages 265–8 of the April on September 25, 2021 by guest. Protected copyright. with autoimmune disorders, three of whom pulmonary function laboratory as well as the issue of Thorax, regret that some errors had CFA.2 As in the case described by anaesthetics department. —SR occurred in the text and in table 3. On page Orchard et al, prognosis in these three 267 the first line of column 1 should have patients was much poorer than that in the read: “...50yearsatentry was 56% patients with high grade pulmonary non- Asthma: Basic Mechanisms and compared with 50% for the subjects over 50 Hodgkin’s lymphomas unassociated with Clinical Management. 3rd Edition. years . . .”. Table 3 is reproduced here with the CFA, presumably due to the combined Barnes PJ, Rodger IW, Thomson NC, eds. corrections shown in bold italics. eVects of the two diseases. ANDREW NICHOLSON BRYAN CORRIN Table 3 Mean (SD) FEV1 % predicted and FEV1/FVC of index and non-index cases stratified by Department of Histopathology, age at entry Royal Brompton Hospital, Sydney Street, London SW3 6NP, p value UK Index cases Non-index cases (t test) All age groups n = 27 n = 40 1 Orchard TR, Eraut CD, Davison AG. Non- FEV1 (% predicted) 54 (25) 100 (21) <0.001 Hodgkin’s lymphoma arising in cryptogenic FEV /FVC 0.57 (0.18) 0.79 (0.13) <0.001 fibrosing alveolitis. Thorax 1998; :228–9. 1 53 N (%) with FEV % pred 70% 20 (74%) 3 (8%) <0.001 2 Nicholson AG, Wotherspoon AC, Jones AL, et 1 < al. Pulmonary B-cell non-Hodgkin’s lymphoma associated with autoimmune Age at entry <50 years n = 8 n = 26 disorders: a clinicopathological review of six FEV1 (% predicted) 56 (37) 100 (19) <0.001 cases. Eur Respir J 1996;9:2022–5. FEV1/FVC 0.53 (0.20) 0.80 (0.12) <0.001 N (%) with FEV1 % pred <70% 4 (50%) 2 (8%) <0.001 AUTHOR’S REPLY We are grateful to Dr Nicholson and Professor Corrin for pointing Ageatentry>50years n=19 n=14 FEV (% predicted) <0.001 out their very interesting report, which was 1 50 (20) 101 (24) FEV1/FVC 0.58 (0.17) 0.78 (0.14) <0.001 published after the original writing of our N (%) with FEV % pred <70% 16 (84%) 1 (7%) <0.001 case report. 1