Study of bronchoalveolar lavage in leptospirosis patients with pulmonary involvement

Abstracts

INTRODUÇÃO: O comprometimento pulmonar é freqüente na leptospirose e caracteriza-se por hemoptise, dispnéia e infiltrados pulmonares bilaterais no radiograma de tórax. Esses achados podem ser compatíveis com hemorragia alveolar, previamente descrita por alguns autores em autópsias e em lavado broncoalveolar. OBJETIVO: Avaliar a presença de hemorragia alveolar, diagnosticada por meio do lavado broncoalveolar, em pacientes portadores de leptospirose com alterações pulmonares, enfatizando-se a importância do método para o diagnóstico precoce da complicação. MÉTODO: Sete pacientes com leptospirose foram submetidos à broncoscopia com lavado broncoalveolar. Todos apresentavam sinais e/ou sintomas respiratórios, e/ou infiltrados no radiograma de tórax, e/ou hipoxemia. A hemorragia alveolar foi definida pelos seguintes achados no lavado: porcentagem de siderófagos e"20%, escore de Golde > 100, e/ou presença de líquido hemorrágico. Foram realizados exame direto e cultura para Leptospiras, com o uso de meios específicos. O diagnóstico da doença foi confirmado por soroaglutinação microscópica para leptospirose. RESULTADOS: O aspecto da broncoscopia foi normal em 5 pacientes, mostrou sangramento na árvore brônquica em 1 caso e sinais inflamatórios em outro. O aspecto do lavado foi hemorrágico em todos os pacientes, configurando o quadro de hemorragia alveolar. A pesquisa direta e a cultura para Leptospiras foram negativas. CONCLUSÃO: A leptospirose deve ser considerada no diagnóstico diferencial das hemorragias alveolares.O lavado broncoalveolar mostrou-se um método eficaz para a detecção de hemorragia alveolar na leptospirose, servindo para orientar a terapêutica imediata, com a finalidade de prevenir sua evolução, caracterizada pela presença de hemoptises maciças e insuficiência respiratória.

Leptospirose; Líquido da lavagem broncoalveolar; Radiografia torácica

ORIGINAL ARTICLE

Study of bronchoalveolar lavage in leptospirosis patients with pulmonary involvement^*

Jorge Eduardo Manhães de Carvalho; Isabela Nascimento Moraes; Angela Santos Ferreira^(TE-SBPT); Regina Lúcia Caetano Gomes; Marcos Olivier Dalston; João José Pereira da Silva

^{Correspondence}

ABSTRACT

BACKGROUND: Pulmonary involvement is common in leptospirosis and usually characterized by hemoptysis, dyspnea and diffuse bilateral infiltrates in chest X-rays. Such findings may be compatible with alveolar hemorrhage, already described by some authors both in autopsies and bronchoalveolar lavage (BAL).

OBJECTIVE: To evaluate the presence of alveolar hemorrhage, diagnosed through BAL, in bearers of leptospirosis patients with pulmonary involvement emphasizing the methodís importance for early detection of this complication.

METHOD: Seven patients with leptospirosis were submitted to BAL. All presented respiratory symptoms and/or infiltrates in the chest X-rays and/or hypoxemia.Alveolar hemorrhage was defined by the following findings in BAL: percentage of siderophages eî20% and/ or Golde score >100 and/or hemorrhagic fluid. Culture and direct tests for leptospirosis were performed in BAL. Diagnosis of disease was confirmed by microscopy serum agglutination.

RESULTS: The aspect of the bronchoscopy was normal in five patients, showed blood in the bronchial tree in one case and inflammatory manifestations in another. The BAL aspect was hemorrhagic for all patients portraying alveolar hemorrhage. Culture and direct tests were negative for Leptospiras in the BAL.

CONCLUSIONS: Leptospirosis must be taken into account in the differential diagnosis of alveolar hemorrhage. The BAL was confirmed as an efficient method for detection of alveolar hemorrhage in leptospirosis, to recommend immediate therapy for the purpose of preventing its evolution to massive hemoptysis and respiratory failure.

Key words:Leptospirosis. Bronchoalveolar lavage fluid. Radiography, thoracic.

Abbreviations used in this paper:

BAL – Bronchoalveolar lavage

BALF – Bronchoalveolar lavage fluid

PaO₂ – Arterial oxygen tension

UFF – Universidade Federal Fluminense

Introduction

Leptospirosis is an acute infectious disease caused by spirochetes of the genus Leptospira, affecting domestic and wild animals in various regions of the world. Humans can occasionally become infected through contact with floodwaters contaminated by rat urine, which is the most common transmission source.⁽²⁾ In human infection, various organs, including the lungs, may be affected.⁽³⁾ Pulmonary involvement is seen in 20% to 70% of leptospirosis patients.^(1,3) Leptospirosis studies conducted in Brazil have described pulmonary involvement that may appear as simple interstitial pneumonia or, in severe cases, massive pulmonary hemorrhage.^(3-5) In leptospirosis cases, pulmonary involvement is clinically characterized by hemoptysis, dyspnea, and pulmonary infiltrates seen in chest X-rays, usually followed by hypoxemia, meeting the criteria for a diagnosis of alveolar hemorrhage.^(6,7) Alveolar hemorrhage from leptospirosis usually appears during the first week of the disease, occasionally in an asymptomatic form,⁽⁸⁾ and typically resolves spontaneously within a few days. However, severe forms of the disease, characterized by rapid evolution to massive hemoptysis, respiratory insufficiency and death from asphyxia, have become more common.^(5,7,9)

The objective of this study was to evaluate, through bronchoalveolar lavage (BAL), alveolar hemorrhage in patients diagnosed with leptospirosis and presenting pulmonary involvement. This was done in an attempt to detect this complication early, thereby facilitating the timely application of appropriate therapeutic measures.

Methods

A prospective study involving 7 hospitalized patients diagnosed with leptospirosis and presenting pulmonary involvement was carried out from January 1999 to February 2002 at the Hospital Universitário Antônio Pedro of the Universidade Federal Fluminense (UFF) in the city of Niterói, in the state of Rio de Janeiro, Brazil. The selection of patients was based on respiratory signs or symptoms presented at admission, including cough, dyspnea, hemoptysis, altered breath sounds, changes in chest X-rays or an arterial oxygen tension (PaO₂) in room air of < 75 mmHg (hypoxemia). Mean age was 40, ranging from 20 to 66. Of the 7 patients, 6 were male and 5 were smokers. All patients reported exposure to rat-infested floodwaters within the 2 weeks prior to the onset of symptoms. Six of the 7 patients were jaundiced, and 2 of those 6 presented renal insufficiency and required hemodialysis. The disease was resolved in 6 patients, and 1 patient suffering from Weil syndrome died of multiple organ failure. In all 7 patients, the diagnosis was confirmed through detection of leptospiral antibodies by microscopic agglutination test. The predominant antibodies were to Leptospira icterohaemorrhagiae serovar copenhageni. Clinical pulmonary symptoms included cough (in 4 patients), hemoptysis (in 4 patients), dyspnea (in 3 patients) and altered breath sounds (in 2 patients). Chest X-rays revealed interstitial infiltrate in 3 patients, alveolar-interstitial infiltrate in 2 patients, alveolar infiltrate in 1 patient, pleural effusion in 1 patient and normal results in 1 patient. Of the 7 patients, 3 underwent computed tomography, and bilateral peripheral ground-glass opacity was seen in all 3. The patient whose chest X-ray was normal presented hemoptysis. Written informed consent was obtained from all patients included in the study, which was approved by the Ethics Committee of the UFF Hospital Antônio Pedro.

On average, all patients underwent fiberoptic bronchoscopy and BAL on the sixth day of hospitalization (from the third to the ninth day). The same professional performed all procedures, using a fiberoptic bronchoscope (Pentax FB 18X^®, Asahi Optical Co., Tokyo, Japan). Due to the diffuse pulmonary involvement revealed in the chest X-rays, the bronchoscope was introduced orally into the segment or subsegment of the middle lobe or into the lingula. Six 20-mL aliquots of sterile saline solution (in a total volume of 120 mL) were infused through the bronchoscope and the solution was then manually aspirated with a 20-mL disposable plastic syringe. The BAL fluid (BALF) was collected into polypropylene tubes, which were placed in an ice bath and immediately sent to the hospital immunopathology laboratory for analysis.

After measuring the total volume of BALF, 20 mL of the liquid were filtered through sterile gauze, and 5 mL were placed in an ice bath and immediately shipped to the Bacteriology Department of the Fundação Oswaldo Cruz in Rio de Janeiro for direct examination and culture for leptospires. The remainder of the BALF was used for laboratory testing (in the laboratory of the Hospital Universitário Antonio Pedro) to detect acid-fast bacilli, fungi and protozoa. The BALF was filtered, placed in 2 plastic centrifuge tubes and centrifuged at 200 xg for 10 minutes in a standard laboratory centrifuge. The supernatant was stored at -80º C. Total cellularity was determined in a Neubauer chamber, and cellular viability was determined by trypan blue exclusion test. The final concentration was adjusted to 2 x 10v cells/mL, and differential counts were performed in Wright-Giemsa stained cytocentrifuge preparations. The cytocentrifuge used was a Shandon Cytospin 2 (Shandon, Tokyo, Japan). At least 200 cells were examined, and percentages of macrophages, lymphocytes, eosinophils, and polymorphonuclear leukocytes were determined. Epithelial cells and erythrocytes were not counted.⁽¹⁰⁾ Perl’s iron stain (Prussian blue), as well as siderophore percentages⁽¹¹⁾ and Golde score,⁽¹²⁾ were used to detect alveolar hemorrhage. The diagnosis of alveolar hemorrhage was confirmed by indications of progressive hemorrhage (in at least BALF 3 samples), by a finding of ≥ 20% siderophores, or by a Golde score > 100.⁽¹¹⁾

Results

Bronchoscopy revealed normal results in 5 patients, tracheobronchial tree hemorrhage in 1 and inflammatory signs in 1. There were no abnormalities during the examination. After bronchoscopy, one patient had fever (37.8°C), which was reduced with antipyretic drugs. The mean volume (all patients) of aspirate recovered was 48 mL of the original 120 mL of sterile saline solution infused. Evidence of hemorrhage was found in the BALF of all patients. The viability of cells in the BALF was within the limits considered acceptable for in vitro cellular functionality studies (86% to 100%). In 6 patients, a significant increase in total cellularity was found, and increased percentages of neutrophils were seen in 3 patients. In 5 patients, the diagnosis of alveolar hemorrhage was confirmed through all three methods (evidence of hemorrhage in the BALF, siderophores > 20% and Golde score >100), and the other 2 patients were diagnosed solely on the basis of evidence of hemorrhage in the BALF. Of the 7 patients, 4 presented thrombocytopenia. Table 1 shows endoscopic and cytologic alterations in the BALF and peripheral blood of the patients studied. For all 7 patients, the results from cultures for acid-fast bacilli, fungi and protozoa, as well as from direct test and culture for leptospires in the BALF, were negative.

Discussion

The BAL procedure has become part of everyday practice and has both diagnostic and therapeutic objectives. It is a simple, safe procedure, and it is quite efficacious in the collection of data regarding upper-airway cells and substances involved in the immunopathogenesis of various respiratory diseases, as well as in the consequent inflammatory lesions.⁽¹³⁾ Recently, BAL has been used for the diagnosis of alveolar hemorrhage of various etiologies.⁽¹⁴⁾ There have been few studies evaluating the use of BAL in cases of leptospirosis.^(8,15) In a recent study conducted in France, Couëdic et al.⁽⁸⁾ used BAL as a diagnostic tool in leptospirosis patients suspected of suffering from alveolar hemorrhage. The authors reported that the condition may appear in an asymptomatic form during the initial stages of pulmonary involvement and highlighted the importance of its early detection so that appropriate therapy can be introduced. If steps are not taken, alveolar hemorrhage may lead to massive hemoptysis and death. For patients in whom the diagnosis of leptospirosis has not been confirmed by serology, patient epidemiological history, as well as clinical and laboratory findings characteristic of leptospirosis, must be taken into consideration before BAL is performed.

Anatomopathological findings described by Arean⁽¹⁶⁾ in a report of 33 fatal leptospirosis cases and by Comby et al.⁽¹⁷⁾ in an autopsy study of 10 cases revealed leptospirosis-induced pulmonary lesions of varying degree, primarily characterized by areas of hemorrhage, edema, capillary congestion, and small foci of infiltrating leukocytes. Pereira da Silva et al.⁽¹⁸⁾ reported extensive areas of edema and alveolar hemorrhage in lung tissue.

In the present study, the endoscopic examination of the tracheobronchial tree of most patients revealed normal results, which indicates that the hemorrhage was, in fact, alveolar. Seijo et al.⁽¹⁹⁾ performed fiberoptic bronchoscopy in 2 patients with leptospirosis who later died of respiratory insufficiency. The bronchoscopy revealed a large amount of blood in the tracheobronchial tree, but no endobronchial lesions. Couëdic et al.⁽⁸⁾ performed BAL in 23 patients with leptospirosis. The examination revealed alveolar hemorrhage in 20 patients, 13 of whom presented pulmonary manifestations. In the present study, the absence of endoscopic lesions (another potential source of bleeding), together with the macroscopic evidence of hemorrhage in the BALF and the large number of erythrocytes seen under optical microscopy, suggests that alveolar hemorrhage was responsible for the pulmonary changes seen. Therefore, we believe that fiberoptic bronchoscopy followed by BAL is an excellent diagnostic protocol for early detection of alveolar hemorrhage in patients with leptospirosis-induced pulmonary changes found during clinical examination, in radiological images, or in blood gas analysis. However, less invasive techniques, such as diffusion capacity for carbon monoxide^(8,14) or computed tomography scans,⁽²⁰⁾ can present viable alternatives. The BAL procedure can also be useful in the differential diagnosis between the severe pulmonary form of leptospirosis and hantavirus pulmonary syndrome. Both diseases have similar presentations, although the BALF of leptospirosis patients is hemorrhagic, a condition not typically seen in cases of hantavirus pulmonary syndrome.⁽²¹⁾

The cause of alveolar hemorrhage in leptospirosis remains unknown. The hypothesis implicating autoimmune disease has not been confirmed.⁽⁸⁾ Leptospiral antigens in lung tissues have been found in bronchial biopsies⁽²²⁾ and in autopsies,^(9,23,24) suggesting a direct action of the microorganism (or of its toxic products) on pulmonary vascular endothelium. Coagulation disorders, such as thrombocytopenia, which is frequent in leptospirosis, seem to act more as a coadjutant than as a direct cause of alveolar hemorrhage.^(23,24) It has recently been suggested that cytokines, in combination with high plasma levels of tumor necrosis factor-alpha, may participate in leptospirosis etiology and contribute to higher rates of severity and mortality.⁽²⁵⁾

Another aspect that should be highlighted is that there were 5 smokers in this study, 4 of whom had positive outcomes and 1 of whom died. The relationship between smoking and the probability of developing pulmonary hemorrhage in some diseases, such as Goodpasture’s syndrome, has previously been described.⁽²⁶⁾ The components of tobacco smoke may increase the permeability of pulmonary capillaries, damaging the alveolar basal membrane and intensifying local inflammatory responses. In a recent study, Garcia et al.⁽²⁷⁾ reported a high incidence of pulmonary involvement in patients with leptospirosis, and smoking (more than 20 cigarettes a day) was found to be a significant risk factor.

In cases of leptospirosis, treatment for alveolar hemorrhage should begin as soon as possible. Oxygen therapy and positive end-expiratory pressure ventilation are the procedures of choice in preventing acute respiratory distress syndrome⁽²⁸⁾ and massive hemoptysis.⁽²⁹⁾ Corticoid therapy has also been used.^(30-32). In a recent study, Trivedi et al.⁽³³⁾ recommend a 3-day course methylprednisolone pulse therapy immediately after the appearance of dyspnea, as a means of impeding the evolution of the process. However, there is no conclusive data on the efficacy of corticosteroids in the prevention and treatment of pulmonary hemorrhage syndrome. It is important to be aware of the extreme severity of pulmonary hemorrhage, which usually appears within the first week, rapidly becomes pronounced and continues to progress.^(5,7,9) This complication should also be considered in the diagnosis of patients who present signs and symptoms compatible with leptospirosis in clinical examinations, as well as in epidemiological and laboratory testing.

Pulmonary signs and symptoms found upon clinical examination, alterations in chest X-rays or in computed tomography scans, serial blood gas analysis indicating hypoxemia (with or without hypocapnia) and blood tests revealing erythrocyte decrease and thrombocytopenia may all be important factors for establishing a diagnosis of alveolar hemorrhage.^(11,14) The BAL procedure has proven to be a valuable tool in the early detection of leptospirosis-induced alveolar hemorrhage,^(8,34) allowing immediate and safer treatment of this severe complication, which is responsible for higher leptospirosis mortality rates.

References

Submitted: 7 July 2003.

Accepted, after revision: 12 November 2003.

• 1. Levett PN. Leptospirosis. Clin Microbiol Rev 2001;14:296-326.
• 2. Feigin RD, Anderson DE. Human leptospirosis. Crit Rev Clin Lab Sci 1975;5:413-67.
• 3. Carvalho JEM. Comprometimento pulmonar na leptospirose [tese]. Rio de Janeiro: Universidade Federal Fluminense; 1989.
• 4. Silva JJP, Paiva LM. Silva JB, Netto BA. Estudo do comprometimento pulmonar na doença de Weil. Rev Inst Med Trop São Paulo 1976;18:387-92.
• 5. Gonçalves AJR, Carvalho JEM, Silva JBG, et al. Hemoptises e síndrome de angústia respiratória aguda como causa de morte na leptospirose. Mudança dos padrões clínicos e anátomo-patológicos. Rev Soc Bras Med Trop 1992;25:261-70.
• 6. Carvalho JEM, Marchiori ES, Silva JBG, et al. Comprometimento pulmonar na leptospirose. Rev Soc Bras Med Trop 1992;25:21-30.
• 7. Silva JJP, Carvalho JEM, Dalston MO, et al. Forma pulmonar grave da leptospirose (FPGL): uma nova apresentação clínica da doença no Estado do Rio de Janeiro.Brasil. Arq Bras Med 1998;72:169-71.
• 8. Du Couëdic L, Courtin JP, Poubeau P, Tanguy B, Di Francia M, Arvin-Berod C. Hemorragies intra-alveolaires patentes et occultes au cours des leptospiroses. Rev Mal Respir 1998;15:61-7.
• 9. Silva JJP, Dalston MO, Carvalho JEM, et al. Clinicopathological and immunohistochemical features of the severe pulmonary form of leptospirosis. Rev Soc Bras Med Trop 2002;35:395-99.
• 10. The Bal Cooperative Group Steering Comitee. Bronchoalveolar lavage constituents in healthy individuals, idiopathic pulmonary fibrosis, and select comparison groups. Am Rev Respir Dis 1990;141(5 Part 2):S169-202.
• 11. De Lassance A, Fleury-Feith J, Escudier E, Beaume J, Bernaudin JF, Cordonnier C. Alveolar hemorrhage. Diagnostic criteria and results in 194 immunocompromised hosts. Am J Respir Crit Care Med 1995;151:157-63.
• 12. Golde DW, Drew WL, Klein HZ, et al. Occult pulmonary hemorrhage in leukemia. Br Med J 1975;2:166-8.
• 13. Reynolds HH. Bronchoalveolar lavage. Am Rev Respir Dis 1987;135:250-63.
• 14. Capron F. Lavage broncho-alvéolaire et hémorragie alvéolaire. Ann Pathol 1999;19:395-400.
• 15. Paganin F, Gaüzere BA, Lugagne N, Blanc P, Roblin X. Bronchoalveolar lavage in rapid diagnosis of leptospirosis. Lancet 1996;347:1562-3.
• 16. Arean VM. The pathologic anatomy and pathogenesis of fatal human leptospirosis (Weil's disease). Am J Pathol 1962;40:393-423.
• 17. Comby F, Gauthier R, Nazinoff O. Nouvelle contribution à l'étude dês leptospiroses à la Réunion II. Anatomopathologie et histopathologie de dix cas mortels. Bull Soc Pathol Exot 1969;62:92-101.
• 18. Silva JJP, et al. Estudo preliminar das leptospiroses no Estado do Rio de Janeiro. Rev Soc Bras Med Trop 1968;2:316.
• 19. Seijo A, Coto H, San Juan J, et al. Distres respiratorio debido a hemorragia pulmonar por leptospirosis. Medicina (Buenos Aires) 2002;62:135-40.
• 20. Marchiori E, Borges IT, Carvalho JEM, et al. Leptospirose forma pulmonar hemorrágica - aspectos na tomografia computadorizada de alta resolução. Rev Imagem 2002;24:83-7.
• 21. Levy H, Simpson SQ. Hantavirus pulmonary syndrome. Am J Respir Crit Care Med 1994;149:1710-3.
• 22. Carvalho JEM, Rodrigues CC, Silva JJP, et al. Identificação da leptospira em tecido pulmonar por broncofibroscopia e biópsia brônquica. Pulmão (RJ) 1999;8:377-81.
• 23. Nicodemo AC, Duarte MIS, Alves AF, Takakura CF, Santos RT, Nicodemo EL. Lung lesions in human leptospirosis: microscopic, immunohistochemical, and ultrastrutural features related to thombocytopenia. Am J Trop Med Hyg 1997;56:181-7.
• 24. Park SK, Lee SH, Rhee YK, Kang SK, Kim KJ, Kim MC, et al. Leptospirosis in Chonbuk Province of Korea in 1987: a study of 93 patients. Am J Trop Med Hyg 1989;41:345-51.
• 25. Tajiki MH, Salomão R. The ratio levels of IL-10/TNF± and its relationship to disease severity and survival in patients with leptospirosis. Braz J Infect Dis 1997;1:138-41.
• 26. Donaghy M, Rees AJ. Cigarette smoking and lung haemorrhage in glomerulonephritis caused by autoantibodies to glomerular basement membrane. Lancet 1983;2:1390-3.
• 27. Martinez Garcia MA, de Diego Damia A, Menendez Villanueva R, Lopez Hontagas JL. Pulmonary involvement in leptospirosis. Eur J Clin Microbiol Infect Dis 2000;19:471-4.
• 28. Weigelt JA, Mitchell RA, Snyder WH. Early positive end-expiratory pressure in the adult respiratory distress syndrome. Arch Surg 1979;114:497-501.
• 29. De Paula A, Assaf M, Monteiro NP, et al. Hemoptises maciças. JBM 1983;44:78-92.
• 30. Courtin JP, Carré PH, Poubeau P, Di Francia M, Jarlet E, Michault A, et al. Hémorragie alvéolaire diffuse et myosite au cours d'une leptospirose ictéro-hémorragique. Contrôle rapide par un seul bolus de corticoides. Rev Mal Respir 1994;11:601-3.
• 31. Burke B, Saerle J, Mattingly D. Leptospirosis presenting with profuse haemoptysis. Br Med J 1976;2:982.
• 32. Kahn JB. A case of Weil's disease requiering steroid therapy for thrombocytemia and bleeding. Am J Trop Med Hyg 1982;31:1213-6.
• 33. Trivedi SV, Chavda RK, Wadia PZ, Sheth V, Bhagade PN, Trivedi SP, et al. The role of glucocorticoid pulse therapy in pulmonary involvement in leptospirosis. J Am Physicians India 2001;49:901-3.
• 34. Carvalho JEM, Ferreira AS, Silva JJP, et al. Hemorragia alveolar na leptospirose diagnóstico pelo lavado broncoalveolar. Pulmão (RJ) 2000;9:27-31.

*

Study carried out at the Hospital Universitário Antonio Pedro, Universidade Federal Fluminense (UFF), Niterói, Rio de Janeiro.

Publication Dates

• Publication in this collection
  08 June 2004
• Date of issue
  Apr 2004

History

• Accepted
  12 Nov 2003
• Received
  07 July 2003