this patient’s CD4’ T-lymphocyte depletion, including HIV infection. Thus, our patient with idiopathic CD4+ T-lymphocytopaenia had suffered from miliary tuberculosis with acute respiratory failure.

Acknowledgement

The authors are indebted to Dr N. Yoshihara, Japan National Institute of Health for testing by Western blot and enzyme immunoassay for antibody against HIV-2.

References

1. Centers for Disease Control. Unexplained CD4+ T-lymphocyte depletion in persons without evident HIV infection - United States. MMWR 1992; 41: 541- 545.

2. Smith DK, Neal JJ, Holnberg SD. Unexplained oppor- tunistic infections and CD4’ T-lymphopenia without HIV infection. N Engl J Med 1993; 328: 373-379.

CASE REPORTS 979

Laurence J. T-cell subsets in health, infectious disease, and idiopathic CD4’ T-lymphocytopenia. Ann Intern Med 1993: 119: 55-62. Beck JS, Potts RC, Kardjito T et al. T4 lymphopenia in patients with active tuberculosis. Clin E.xp Imrnonol 1985; 60: 49-54.

Greenberg SD, Frager D, Suster B et al. Active pulmon- ary tuberculosis in patients with AIDS: spectrum of radiographic findings (including a normal appearance). Radiology 1994; 193: 116-119. Japan Health and Welfare Statistics Association. Health and Welfare Statistics, 1995 Tokyo, 1995 (in Japanese); Tokyo. Turett GS, Telzak EE. Normalization of CD4’ T- lymphocyte depletion in patients without HIV infection treated for tuberculosis. Chest 1994; 105: 1335-1337. Ho DD, Cao Y, Zhu T et al. Idiopathic CD4+ T- lymphocytopenia - immunodeficiency without evidence of HIV infection. N Engl J Med 1993; 328: 380-385. Soriano V, Hewlett, I, Heredia A et al. Idiopathic CD4+ T-lymphocytopenia. Lancer 1992; 340: 607-608.

Diagnosis of pulmonary veno-occlusive disease: new criteria for biopsy

L. VALD&*, J. R. GONZALEZ-JUANATEY~, D. ALVAREZ”, J. ANT~NEZ*, J. MANUEL VALLE”, P. PENELA* AND R. ALVAREZ-SALA’

“Seccidn de Neumologia, Hospital de Conxo ‘Servicio de Cardiologia, Hospital Xeral de Galicia *Servicio de Anatomia Patolbgica, Hospital Xeral de Galicia, Complexo Hospitalario Universitario de Santiago, Santiago de Compostela, Spain ‘Servicio de Neumologia, Hospital Universitario La Paz, Madrid, Spain

Introduction

Pulmonary veno-occlusive disease (PVOD) is a rare condi- tion in which the predominant anomaly is the stenosis or occlusion of the lumen of small pulmonary veins due to fibrosis of the intima (1). Its aetiology is unknown, although it is sometimes associated with viral infection, environmental toxins, chemotherapy, autoimmune disease, use of contraceptives, intracardiac shunts or radiation injury, and some cases suggest genetic predisposition (2). In

Received 10 March 1997 and accepted in revised form 27 June 1997. Correspondence should be addressed to: L. ValdCs, C/. Emilia Pardo BazAn 26, Urbanizacibn parque Montouto, 15894 Teo, La Coruiia, Spain.

spite of a variety of therapies having been tried, it is usually fatal within a few years; lung transplant is currently the treatment of choice. Its definitive diagnosis requires dem- onstration of the above-mentioned anatomopathological features in pulmonary biopsy material, although pulmon- ary biopsy is not always possible. The diagnostic difficulties associated with three cases of PVOD seen in our centre in recent years have led us to examine what clinical criteria constitute sufficient grounds for carrying out pulmonary biopsy to confirm PVOD.

Case Reports

For all three patients, cardiac and pulmonary circulation parameters measured by cardiac catheterization are listed in Table 1, and pulmonary function parameters in Table 2.

980 CASE REPORTS

TABLE 1. Haemodynamic parameters of three patients with pulmonary veno-occlusive disease

Case 1 Case 2 Case 3

RA (mmHg) 28 14 2 RV (mmHg) 94 75 68 Pulmonary artery (mmHg) 94147165 1513Ol45 68/23/38 Pulmonary capillary wedge pressure (mmHg) 35 20 16 LV (mmHg) 13514 14015 CO (lmin ~ ‘) 3.2 6.45 CI (lminm - *) 2.38 3.83 Total pulmonary resistance (dyn s - ’ cm - 5, 1625 471

RA, Right atrium; RV, right ventricle; LV, left ventricle, CO, cardiac output; CI, cardiac index.

TABLE 2. Pulmonary function parameters of three patients with pulmonary veno-occlusive disease, with expression as a percentage of the theoretical value in parentheses

Case 1 Case 2 Case 3

FVC (ml) 2180 (94%) 2050 (62%) 2400 (62%) FEV, 1600 (98%) 1230 (49%) 1360 (46%) FEV, % 73 (70) 60 (75) 56 (76) D LCO (“4 86 22 102 D/VA (%) 79 29 127

CASE 1

A 69-year-old woman was admitted for investigation of dyspnoea that had developed over several years, was currently experienced at rest and was accompanied by orthopnoea and swollen lower extremities. Physical exami- nation revealed central and peripheral cyanosis, jugular dilation at 45”, and fovea1 malleolar oedemas. Blood bio- chemistry and haematology were normal. Chest radiogra- phy and computed tomography (CT) scans showed cardiomegaly and prominent pulmonary hila, and the CT scans also showed smooth interlobular septal thickening and ground glass opacity. Arterial blood gas parameters at admission were as follows: PaO,, 57 mmHG; PaCO,, 31.3 mmHg; and pH 7.46 breathing room air. Enlargement of right heart cavities was suggested by electrocardiography and confirmed by echocardiography, which also showed tricuspid insufficiency and severe pulmonary artery hyper- tension (PAHT). Left ventricular systolic and diastolic function were normal. Pulmonary perfusion gammagraphy results were normal. Pulmonary angiography showed a poor peripheral vascular tree on both sides, with diffuse stenosis of the pulmonary veins. Lower limb venography results were normal. In lingular biopsy material obtained by minithoracotomy, venous lumina were partially occluded due to fibrosis of the intima and associated recanalized thrombi (Plate 1); PVOD was diagnosed. The biopsy material also showed alterations typical of PAHT (reduced arterial lumina due to fibrosis of the intima).

PLATE 1. A pulmonary vein with a partially recanalized thrombus. (Haematoxylin-eosin x 100).

CASE 2

A 58-year-old man admitted on account of dyspnoea had been hospitalized on numerous occasions since diagnosis of chronic obstructive pulmonary disease (COPD) 4 years previously. Prior to diagnosis of COPD he had smoked 20 cigarettes a day; since then he had suffered dyspnoea in response to mild effort and had had home oxygen treat- ment. In the days preceding his latest admission he had an irritant cough, dyspnoea at rest, and swollen lower extremities. Physical examination revealed obesity, jugular dilation at 45”, disseminate rhonci with basal crackle on both sides, and fovea1 malleolar oedemas. Blood biochem- istry and haematology were normal. Chest radiography showed cardiomegaly and bilateral alveolar pattern, which responded to diuretic therapy and oxygen (Plate 2). Chest CT scans showed smooth interlobular septal thickening, ground glass opacity, enlarged central pulmonary arteries and pulmonary veins of normal caliber. Arterial blood gas parameters at admission were as follows: PaO,, 53 mmHG; PaCO,, 40 mmHg; pH 7.34. Enlargement of right heart cavities was suggested by electrocardiography and con- firmed by echocardiography, which also showed normal left ventricular structure and systolic and diastolic function. Pulmonary perfusion gammagraphy showed hypoperfused

CASE REPORTS 981

PLATE 2. Postero-anterior chest radiographs of the Case 2 patient, showing (a) cardiomegaly and bilateral alveolar pattern, and (b) improvement of same after treatment with diuretics and oxygen.

regions in both lung bases. Pleural fluid biochemistry was typical of a transudate. The apnoea/hypoapnoea rate in a polysomnogram was three per h. In ligular biopsy material obtained by minithoracotomy, pulmonary venous lumina were partially occluded due to fibrosis of the intima and fibrous thickening of the wall. PVOD was diagnosed.

CASE 3

A 5 1 -year-old man who smoked 20 cigarettes a day and had a daily alcohol intake of 120 g was admitted on account of dyspnoea. Bronchial asthma had been diagnosed 5 years previously, and he currently suffered exertional dyspnoea in response to mild effort and had home oxygen treatment. In the days preceding admission he had had dyspnoea at rest, paroxysmal nocturnal dyspnoea, cough and purulent spu- tum. Physical examination revealed central and peripheral cyanosis and jugular dilation at 45”; auscultation revealed greatly diminished ventilation and disseminate rhonci and wheezing. Blood biochemistry and haematology were nor- mal. Chest radiography showed cardiomegaly, Kerley B lines and bilateral pleural effusion, which responded to diuretic therapy and oxygen. Chest CT scans showed

smooth interlobular septal thickening, enlarged central pulmonary arteries, pulmonary veins of normal caliber and bilateral pleural effusion. Arterial blood gas parameters at admission were as follows: PaO,, 35 mmHg; PaCO, 83 mmHg; pH 7.33; bicarbonate 44 mmol l- ‘; BE 18 mmol l- ‘; SaO,, 60%. Enlargement of right heart cavities was suggested by electrocardiography and con- firmed by echocardiography, which also showed tricuspid insufficiency and normal left ventricular structure and systolic and diastolic function. Pulmonary perfusion gammagraphy results were normal. In pulmonary biopsy material obtained by minithoracotomy, venous lumina were partially occluded due to fibrosis of the intima and fibrous thickening of the wall. PVOD was diagnosed.

Discussion

PVOD is a rare cause of PAHT which usually, although by no means exclusively, affects children or young adults. Its aetiology is unknown. Its definitive diagnosis requires histological proof of extensive occlusion of small pul- monary veins due to fibrosis and proliferation of the intima (1). However, since the clinical situation of patients with this almost invariably fatal disease is generally critical, the performance of biopsy is only justified if clinical find- ings obtained non-invasively suggest PVOD with high probability.

PVOD patients generally have a history of progressive dyspnoea, which in some cases is associated with orthop- noea, paroxysmal nocturnal dyspnoea or syncope. Clinical signs of right cardiac insufficiency may be present, and inspiratory crackles due to pulmonary congestion are often audible upon auscultation of the chest. Chest radiography shows a prominent right ventricle, dilated pulmonary arteries and Kerley’s B lines. Electrocardiograms suggest enlargement of the right ventricle and/or deviation of the axis to the right, while echocardiography shows increased right ventricular systolic pressure and the absence of left cardiac lesions such as pulmonary vein stenosis, mitral stenosis or car triatrium. Haemodynamic measurements show PAHT (with increased pressure in both the pulmon- ary artery and the right ventricle); but pulmonary capillary wedge pressure may be either high or normal (3). Left atria1 pressures are normal.

All three of our patients presented the above features. Although older than is usual for reported PVOD patients, their ages did not exclude PVOD: several authors have reported isolated cases of PVOD patients aged more than 50 years (2,4-7), and all three of the patients described by Palevsky et al. (3) had ages similar to those of our patients. In fact, it seems plausible that PVOD may be considerably more prevalent among mature and elderly adults than is commonly thought, given the ease of misdiagnosis as COPD when advanced age coincides with factors such as smoking.

In view of the characteristics described above, it has been suggested (4) that when pulmonary biopsy is not possible, PVOD may be diagnosed on the basis of the coincidence of severe PAHT, radiologically demonstrated pulmonary

982 CASE REPORTS

congestion and normal pulmonary capillary wedge pres- sures. However, these three signs can also coincide in cases of atria1 mixoma, in which the patient may have intermit- tent pulmonary oedema during episodes of mitral obstruc- tion but show normal pulmonary artery wedge pressure when there is no obstruction. Furthermore, as already noted, pulmonary capillary wedge pressure can be normal in some PVOD cases. We have accordingly reconsidered the effects of PVOD in order to define a modified set of criteria which can be determined without resorting to any invasive procedure and whose fulfilment constitutes sufficient grounds for considering it desirable to perform pulmonary biopsy to confirm PVOD.

Absence of PAHT clearly rules out PVOD. PAHT can almost always be detected by Doppler echocardiography: almost all patients with PAHT exhibit tricuspid insuffi- ciency due to the structural and functional alterations of the right ventricle secondary to the increased right ventricular afterload caused by PAHT (8) and continuous wave Doppler measurements of the maximum tricuspid regurgi- tation velocity allow calculation of the systolic pressure in the pulmonary artery [correlation with values measured by means of pulmonary artery catheterization is excellent (9)]. Furthermore, pulsed Doppler monitoring of pulmonary artery flow velocity clearly differentiates normal pulmonary artery pressures from PAHT; the former produce dome-like velocity profiles with a sharp peak in early systole and decreased acceleration time (10).

The high pressures in the pulmonary veins and capillaries of PVOD patients cause the appearance of radiological signs of postcapillary hypertension and pulmonary conges- tion, and the absence of these signs rules out PVOD. In their presence, the absence of left atria1 dilation differenti- ates PVOD from congestive heart failure (11). In addition, PVOD causes no redistribution of blood flow to higher regions of the lung, whereas mitral stenosis causes redistri- bution of circulation and interstitial and alveolar oedema.

For diagnosis of PVOD, alteration of left cardiac func- tion must be ruled out as a cause of observed pulmonary hypertension. Normal pulmonary capillary wedge pressure is an indirect and imperfect indication of the absence of left cardiac lesions: the wedged catheter senses the pressure in the large pulmonary veins, which are generally unaffected by PVOD (4); but if the large veins are affected, then pulmonary capillary wedge pressure can be above normal, which in the absence of other information would suggest left ventricular alterations rather than PVOD. It is accord- ingly desirable to rule out alterations of the left heart by a more direct technique such as echocardiography. M mode, two-dimensional and Doppler echocardiography together allow accurate evaluation of the structure and function of both cavities and of the mitral and aortic valves (12,13); in particular, the existence of hypertrophy and/or dilation can be investigated directly, as can the existence of valvular stenosis and/or regurgitation, and transvalvular pressure drops and valve orifice areas can be quantified.

Another ultrasound technique that might assist diagnosis of PVOD when the large pulmonary veins are affected (although we know of no case in which it has been used for this purpose) is transoesophageal pulsed Doppler echo-

cardiography. Although usually employed for diagnosis of cardiopathies affecting the left heart, this technique has also revealed stenosis of the large pulmonary veins, both by direct visualization of the stenosis and by demonstrating the consequent accelerated flow (14). However, this technique seems unlikely to be of much use if PVOD affects only venules and small veins, in which case it would probably show just non-specific acceleration at the entry to the left atrium.

Greater potential for diagnosis of PVOD appears to be shown by chest CT. Swensen et al. (15) recently found that in seven out of eight PVOD cases CT scans showed thickening of interlobular septa that correlated with biopsy findings of septal fibrosis and associated venous sclerosis, and that in all eight cases the CT scans showed regions of ground glass opacity (possibly due to the thickening of alveolar septa and associated hyperplasia of lining epi- thelium). Swensen et al. (16) suggest that no disease other than PVOD produces these CT signs in conjunction with CT signs of enlarged central pulmonary arteries, pulmonary veins of normal calibre and pleural effusion. The CT scans of our three patients showed thickened interlobular septa, enlarged central pulmonary arteries and pulmonary veins of normal calibre, but only two showed ground glass opacity and only one pleural effusion. Clearly, consideration of a larger number of cases will be necessary for proper assess- ment of the diagnostic value of these CT signs. Spiral CT scans and magnetic resonance (MR) images may also prove useful for diagnosis of PVOD, although we know of no case in which these techniques have been used for this purpose.

In conclusion, we suggest that a tentative diagnosis of PVOD sufficiently well-founded to justify confirmatory pulmonary biopsy can be achieved by non-invasive methods (basically chest radiography and conventional and Doppler echocardiography), the relevant diagnostic pattern being the coincidence of severe PAHT, radiological evidence of pulmonary oedema and the absence of any alteration of left cardiac structure or function. Unlike previous criteria (4), these conditions rule out the existence of atria1 myxoma. Future studies should establish the value of high resolution CT, spiral CT, MR imaging and trans- oesophageal echocardiography for noninvasive diagnosis of PVOD.

References

1. Bjornsson J, Edward WD. Primary pulmonary hyper- tension: a histopathologic study of 80 cases. Mayo Clin Proc 1985; 60: 16-25.

2. Wagenvoort CA, Wagenvoort N, Takahashi T. Pulmonary veno-occlusive disease; Involvement of pulmonary arteries and review of the literature. Hum Path01 1985; 16: 1033-1041.

3. Palevsky HI, Pietra GG, Fishman AP. Pulmonary veno-occlusive disease and its response to vasodilator agents. Am Rev Respir Dis 1990; 142: 426429.

4. Case Records of the Massachusetts General Hospital (Case 14-1983). N Engl J Med 1983; 308: 823-834.

5. Okubo S, Yutani C, Horimoto M et ul. Pulmonary veno-occlusive disease in elderly man: case report and review of the literature. Jpn Circ J 1984; 48: 602-609.

6. Case Records of the Massachusetts General Hospital (Case 21-1986). N Engl J Med 1986; 314: 1435-1445.

7. Leinonen H, Pohjola-Sintonen S, Krogerus L. Pulmon- ary veno-occlusive disease. Acta Med Stand 1987; 221: 307-3 10.

8. Schiller NB, Himelman R. Echocardiography and Doppler in clinical cardiology. In: Chaterjee K, Cheitlin MD, Karlinev Y, Parmley WW, Rapaport E, Scheinman M, eds. Cardiology: An Illustrated Text/ Reference. New York: Gower Medical Publishing, 1991: Chapter 4: 33-106.

9. Veyrat C, Kalmanson D, Farjou M, Manin JP, Abitbol G. Non-invasive diagnosis and assessment of tricuspid regurgitation and stenosis using one and two- dimensional echopulsed Doppler. Br Heart J 1982; 47: 596-605.

CASE REPORTS 983

10. Naeije R, Torbicki A. More on the noninvasive diag- nosis of pulmonary hypertension: Doppler echocardi- ography revisited. Eur Respir J 1995; 8: 1445-1449.

11. Fraser RG, ParC JAP, Par& PD, Fraser RS, Genereux FP (eds). Pulmonary hypertension and edema. In: Diug- nosis of Diseases of the Chest, 3rd edn. Philadelphia, PA: W.B. Saunders Company, 1990; 1823-1968.

12. Popp RL. Review article. Echocardiography 1 part. ‘-N Engl J Med 1990; 323: 101-09.

13. Popp RL. Review article. Echocardiography 2 part. N Engl J Med 1990; 323: 165-172.

14. Obeid AI, Carlson RJ. Evaluation of pulmonary veins stenosis by transoesophageal echocardiography. J Am Sot Echocardiogr 1995; 8: 888-896.

15. Swensen SJ, Tashjian JH, Myers JL et al. Pulmonary venoocclusive disease; CT findings in eight patients. Am J Radio1 1996; 167: 937-940.

16. Swensen SJ, Aughenbaugh GL, Douglas WW, Myers JL. High-resolution CT of the lungs: findings in various pulmonary diseases. Am J Radio1 1992; 158: 971-979.

Transient cortical blindness: a complication of bronchial artery embolization

S.-F. LIU”, T.-Y. LEE+, S.-L. WONG*, Y.-F. LAI’ AND A.-S. LIN*

“Division of Chest, Department of Internal Medicine and ‘Division of Radiology, Chang Gung Memorial Hospital, Kaohsiung, Taiwan, R.O.C.

Introduction

Bronchial artery embolization is an effective therapeutic alternative for the treatment of severe haemoptysis, especially when conservative treatments fail or when patients are not good candidates for surgery (1). Some complications of bronchial artery embolization have been reported in the literature including chest pain, ectopic deposition of coil and embolization of other vessels, left main bronchial stenosis or infarction (2), bronchoesopha- geal fistula (3), spinal cord injury (4), fatal ischaemic colitis (5), and transient pulmonary infarction after complete pulmonary artery and bronchial artery embolization (6). There is even one report in the literature of left bronchial- to-coronary artery communication, seen on a follow-up postembolization angiogram (7), thus the potential for myocardial infarction with bronchial embolization also

Received 26 May 1997 and accepted in revised form 4 August 1997. Correspondence should be addressed to: S.-F. Liu, Division of Chest, Department of Internal Medicine, Chang Gung Memorial Hospital, Kaohsiung, 123, Ta-Pei Road, Niao-Sung Hsiang, Kaohsiung Hsien, Taiwan, R.O.C.

exists. To our knowledge, this is the first report of cortical blindness following bronchial artery embolization in the English literature.

Case Report

CASE 1

A 29-year-old male was admitted to our hospital emergency room because of persistent haemoptysis for 2 days. He had a history of pulmonary tuberculosis and underwent left lower lobe lobectomy for bronchiectasis in 1990, as well as a bronchial artery embolization for recurrent haemoptysis in 1991. Because of continued haemoptysis after conserva- tive treatments during this hospitalization, a secondary bronchial artery embolization was performed. Angiogram through bilateral bronchial arteries injection via right femoral artery approach revealed dilated and tortuous hypervascularization over the bilateral lung field, with blood supplied from the left inferior and right bronchial artery [Plate l(a) and (b)]. Embolization was performed with ivalon particles (250-590,~m) in 20 ml lipiodol injec- tion in both left inferior artery and right bronchial artery.