D. AngoulvantI. MohammediS. DuperretP.Bouletreau

Septic shock caused byMycobacterium tuberculosisin a non-HIV patient

Received: 9 September 1998Accepted: 12 October 1998

Sir: Septic shock due to Mycobacterium tu-berculosis is rarely reported. Most casesconcern patients with proven HIV infectionor who are at risk of it but have unknownserologic status [1±5]. We report a case ofrefractory septic shock caused by M. tuber-culosis in a non-HIV patient leading tomultiple organ failure and death.

A 44-year-old male was admitted to ourintensive care unit with a combination ofprofound asthenia, anorexia, weight loss,acute respiratory failure, and septic shock.His past medical history included inferiormyocardial infarction, tobacco and alcoholabuse, and cancer of the right tonsil con-sidered cured 2 years previously. On ad-mission, the physical examination revealedthe following vital signs: temperature39.5�C, respiratory rate 22 breaths/min,blood pressure 80/45 mm Hg, and pulse100 beats/min, after volume expansion andunder dopamine (8 mg/kg per min) and do-butamine (5 mg/kg per min) infusion. Cre-pitant rales were found bilaterally. Thepatient was alert and oriented. Chestradiograph showed diffuse bilateral pul-monary infiltrates consistent with ARDS.Arterial blood gases obtained under oxy-gen therapy were pH 7.46, arterial carbondioxide tension 6.3 kPa, arterial oxygentension 7 kPa, and oxygen saturation 87 %.Lactic acid was 3.2 mmol/l. His white bloodcell count was 2400/mm3, hemoglobin 9.9 g/dl, and platelet count 96 000/mm3. Cardiactissue enzymes were normal. The enzyme-linked immunosorbent assay test for HIV 1and 2 was negative. The patient was intu-

bated, and mechanical ventilation waspromptly instituted because of refractoryhypoxemia. A two-dimensional Dopplertransthoracic echocardiography showedleft ventricular depressed systolic functionwithout any indication for a preload in-crease, consistent with severe septic shock.Then the dobutamine dose was increasedto 20 mg/kg per min and norepinephrine in-fusion was started due to persistent hypo-tension. Specimens of urine, blood, andbronchial secretions were sent for culture,and empirical antibiotic treatment consist-ing of ceftriaxone, ofloxacin, and antitu-berculosis drugs (rifampin, isoniazid, andethambutol) was begun. Direct examina-tion of bronchoalveolar lavage revealednumerous acid-fast bacilli, subsequentlyidentified as M. tuberculosis susceptible tothe drugs given. Routine blood and urinecultures were negative. Despite antituber-culous therapy, mechanical ventilation, andmaximum use of catecholamine combina-tion therapy, his condition continued toworsen and he died on day 3 in multipleorgan failure. No autopsy was performed.

To date, nine well-documented cases ofseptic shock due to M. tuberculosis havebeen reported in the English literature. Tu-mor necrosis factor production from myco-bacterial products is assumed to beresponsible for septic shock hemodynamicalteration [2]. Of these nine cases, sevenwere HIV-infected patients [1, 3±5], andtwo were considered to be at high risk forHIV disease, although testing had not beenperformed [2]. It is well known that adultsinfected by HIV have increased suscept-ibility to M. tuberculosis and progress morerapidly to disease. Conversely, the casepresented herein is one of the first reportedof hemodynamic septic shock due to M. tu-berculosis in a patient without HIV. Be-cause two-dimensional echocardiographyis considered a reliable technique to evalu-ate left ventricle preload and contractilityin septic shock [6], we did not used pul-monary artery catheterization to monitorour patient's hemodynamic profile. For thispatient, with a previous history of cancer,several weeks of weight loss, anorexia,asthenia, combined with respiratory symp-toms and chest radiographic aspects on ad-

mission led to the diagnosis. However,despite initiation of prompt and appropri-ate empirical antibiotic therapy and maxi-mum supportive therapy, the infectionproved refractory and the patient died.

We conclude that M. tuberculosisshould be considered in the differential di-agnosis of septic shock, even in non-HIVpatients, especially in patients with othercauses of immunosuppression such as can-cer or transplantation.

References

1. Gachot B, Wolff M, Clair B, RØgnier B(1990) Severe tuberculosis in patientswith human immunodeficiency virus in-fection. Intensive Care Med 16: 491±493

2. Ahuja SS, Ahuja SK, Phelps KR, Thel-mo W, Hill AR (1992) Hemodynamicconfirmation of septic shock in dissemi-nated tuberculosis. Crit Care Med 20:901±903

3. Vadillo M, Corbella X, Carratala J(1994) AIDS presenting as septic shockcaused by Mycobacterium tuberculosis.Scand J Infect Dis 26: 105±106

4. George S, Papa L, Sheils L, MagnussenCR (1996) Septic shock due to dissemi-nated tuberculosis. Clin Infect Dis 22:188±189

5. Clark TM, Burman WJ, Cohn DL, Meh-ler PS (1998) Septic shock from Myco-bacterium tuberculosis after therapy forPneumocystis carinii. Arch Intern Med158: 1033±1035

6. Jardin F, Valtier B, Beauchet A, Du-bourg O, Bourdarias JP (1994) Invasivemonitoring combined with two-dimen-sional echocardiographic study in septicshock. Intensive Care Med 20: 550±554

D. Angoulvant × I.Mohammedi ()) ×S.Duperret × P.BouletreauDepartment of Intensive Care Medicine,Pavillon N, Hopital Edouard Herriot,1 Place d'Arsonval, F-69003 Lyon, Francee-mail: ismael.mohammedi @ chu-lyon.frTel. + 33(04)72 1100 15Fax + 33(04)72 111096

238