Coxiella burnetii Blood Cultures from Acute and Chronic Q-Fever ...

7 downloads 82 Views 173KB Size Report
One patient with acute Q fever had a positive blood culture 25 days after the discontinuation of ... presentations of chronic Q fever are blood culture-negative.
JOURNAL OF CLINICAL MICROBIOLOGY, Dec. 1995, p. 3129–3132 0095-1137/95/$04.0010 Copyright q 1995, American Society for Microbiology

Vol. 33, No. 12

Coxiella burnetii Blood Cultures from Acute and Chronic Q-Fever Patients DIDIER MUSSO

AND

DIDIER RAOULT*

Unite´ des Rickettsies, CNRS EP J0054, Faculte´ de Me´decine, 13385 Marseille Cedex 05, France Received 19 May 1995/Returned for modification 10 August 1995/Accepted 20 September 1995

Q fever is a worldwide zoonosis caused by Coxiella burnetii, an obligate intracellular bacterium living in the phagolysosome of the host cell. The most important reservoirs of this infection are sheep, cattle, and goats. Infection of humans usually occurs by inhalation of infected particles. Two forms of the disease, acute and chronic, are recognized. The most common clinical presentations of acute Q fever are fever, pneumonia, granulomatous hepatitis, and meningoencephalitis (7, 18), although acute Q fever may be asymptomatic (3). The most common presentations of chronic Q fever are blood culture-negative endocarditis, infections of vascular aneurysms or prostheses, and osteomyelitis (1). Laboratory diagnosis of Q fever is usually performed by serological methods relying on microimmunofluorescence, complement fixation, or an enzyme immunoassay (9). C. burnetii is characterized by an antigenic phase variation; phase I organisms are isolated from infected humans or animals, whereas phase II organisms are obtained after serial passages of C. burnetii in embryonated eggs or cell culture systems (4). Acute Q fever is characterized by high levels of anti-phase II antibodies, whereas chronic Q fever is characterized by increasing titers of both anti-phase I and anti-phase II antibodies. Isolation of C. burnetii was first achieved with embryonated eggs (17) and with inoculated animals (21). We have recently developed a cell culture system which uses a shell-vial assay for the isolation of C. burnetii (14) and have used this system to successfully isolate the organism from the blood, bone marrow aspirates, kidneys, and placentas (11) of patients with acute Q fever and from the cardiac valves (14), vascular grafts, and aneurysms (8) of patients with chronic Q fever. Between January 1991 and August 1994, we attempted to isolate C. burnetii from 844 blood samples from patients with suspected cases of Q fever. The results of this study are reported together with corresponding available data on the clinical presentations, antibiotic prescriptions, and serological and blood culture results for the patients.

MATERIALS AND METHODS Patients and clinical samples. Our center, located in Marseille (southern France), is the National Reference Center for Rickettsiosis. The initial diagnosis of Q fever was made by antibody estimation with submitted serum samples. When testing revealed specific anti-C. burnetii antibodies, physicians were asked to provide clinical data together with heparinized blood samples and eventually tissue samples for cell culture. Most Q-fever patients were examined by one of us (D.R.). For the evaluation of blood culture results, only antibiotics capable of inhibiting C. burnetii were considered: doxycycline, quinolones, macrolides, and chloramphenicol (13, 22). For blood culture, a 5-ml sample of heparinized blood was obtained. Each blood specimen was sedimented for 1 h, and 500 ml of the supernatants was collected to inoculate into shell vials. Serological and culture procedures were performed daily. Culture procedure. Isolation of C. burnetii by cell culture was performed as previously reported (14–16). Briefly, culturing was performed by the shell-vial assay with human embryonic lung fibroblasts. Three confluent shell vials were inoculated for each blood sample and were incubated for 6 and 14 days at 378C under a 5% CO2 atmosphere. Detection of C. burnetii was carried out directly with the shell vial by direct immunofluorescence with a locally produced anti-C. burnetii monoclonal antibody (23). If immunofluorescence was negative after 6 and 14 days of incubation, the culture was reported as negative. If immunofluorescence was positive after 6 and 14 days of incubation, the culture was reported as positive. Serological procedures. The immunofluorescence assay was performed as previously described (19). All sera were diluted at 1/25, 1/50, and 1/100 and were screened for total immunoglobulins (Ig). If the screening was positive, the final titers of IgG, IgA, and IgM were determined for both phase I and phase II antibodies. Our criteria for serological diagnosis were anti-phase II IgG antibodies of $1/200 and anti-phase II IgM antibodies of $1/50 for acute Q fever (positive predictive value, 100%) and anti-phase I IgG antibodies of $1/1,600 for chronic Q fever (positive predictive value, 100%) (19).

RESULTS From January 1991 to August 1994, we received 844 blood samples from 591 patients for the isolation of C. burnetii; 551 (65%) samples were from our region (southeastern France) and 293 (35%) samples were from other laboratories. Clinical and serological evidence of Q fever was found in 249 of the patients: 178 had acute Q fever and 71 had chronic Q fever. Of the patients with acute Q fever, 160 (90%) had fever, 124 (70%) had hepatitis, 80 (45%) had pneumonia, and 45 (25%) had all three syndromes. Of the patients with chronic Q fever, 50 (70%) had endocarditis, and the remainder had infections of vascular aneurysms or prostheses. The results of the attempted cultures are presented in Table 1. Acute Q fever. We performed 226 blood cultures for the 178

* Corresponding author. Mailing address: Unite´ des Rickettsies, Faculte´ de Me´decine, 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France. Phone: 33 91 38 55 17. Fax: 33 91 83 03 90. 3129

Downloaded from http://jcm.asm.org/ on December 14, 2017 by guest

Q fever, a worldwide zoonosis caused by Coxiella burnetii, may present as either an acute or a chronic disease. We correlated the results of 844 C. burnetii blood cultures with serological, clinical, and therapeutic data. C. burnetii was isolated from 17% of untreated patients with acute Q fever and from 53% of untreated patients with chronic Q fever. C. burnetii was not isolated from patients who were receiving antibiotics active against C. burnetii. For seven culture-positive patients with acute Q fever, serology was negative when C. burnetii was isolated. One patient with acute Q fever had a positive blood culture 25 days after the discontinuation of specific antibiotic therapy, and another had a positive blood culture after the resolution of symptoms. In one case of chronic Q fever, a positive blood culture resulted from noncompliance with treatment. The culture method described in this report is suitable for all laboratories with cell culture facilities. Our findings suggest that blood samples must be collected prior to the initiation of an antibiotic regimen if C. burnetii is to be successfully isolated.

3130

MUSSO AND RAOULT

J. CLIN. MICROBIOL.

TABLE 1. Blood cultures of C. burnetii for Q-fever patients Condition (no. of patients [no. of cultures])

No. of cultures (no. of patients)

No. of positive cultures

During Before During Before treatmenta treatment treatment treatment

Acute Q fever (178 [226]) Chronic Q fever (71 [162])

66 (66) 17 (17)

160 (116) 145 (71)

11 9

0 0

a Treatment was with antibiotics capable of inhibiting C. burnetii (doxycycline, quinolones, macrolides, and chloramphenicol).

DISCUSSION Positive cultures of C. burnetii were obtained for 17% of the untreated acute Q-fever patients. Among the acute Q-fever patients whose blood yielded C. burnetii, the incidence of specific clinical symptoms was approximately the same as that for all acute Q-fever patients diagnosed in this study. Thus, there is no evidence that blood culture positivity is linked to a particular clinical syndrome. The isolation success rate for C. burnetii for acute Q-fever patients approximates that achieved for the acute phases of other illnesses. For example, in acute Streptococcus pneumoniae pneumonia, the isolation rate ranges from 25 to 35% (2). The rate of 17% found in this study may in fact be lower than that which could be potentially achieved, because in each case, only one blood sample was collected during the bacteremic phase.

TABLE 2. Clinical, serological, therapeutic, and culture data for acute Q-fever patients with blood cultures positive for C. burnetii Patient

Admission diagnosis

Fevera

Hepatitisb

Pneumoniac

1

Treatment (duration)d

1A

Acute hepatitis

1

1

2A

Suspicion of malaria

1

1

3A

Suspicion of malaria

1

1

4A

Surgery (endartectomy)

5A

Febrile illness

1

1

Doxycycline (from day 18)

6A

Suspicion of acute cholecystis

1

1

Doxycycline (from day 13)

7A

Suspicion of malaria

1

1

Doxycycline (day 4 to 25)

8A

Suspicion of meningitis

1

9A 10A

Febrile illness Febrile cutaneous rash

1 1

a

Doxycycline (day 3 to 24) Doxycycline (day 7 to 28)

1

Doxycycline (day 25 to 46) Doxycycline (day 15 to 35)

Doxycycline (day 6 to 27) 1

Erythromycin (day 15 to 22) Rifampin (from day 10)

Anti-phase II IgG and IgM antibody titer (day)

0.0 (3) 0.0 (16) 0.0 (7) 1,600.50 (30) 0.0 (6) 100.25 (18) 400.200 (15) 200.25 (45) 200.100 (60) 0.0 (8) 800.0 (17) 200.100 (11) 400.200 (26) 0.0 (4) 0.0 (11) 0.0 (1) 50.0 (7) 200.0 (15) 0.0 (6)

Blood culture (day)e

1 (3) 1 (7) 1 (6) 2 (18) 2 (15) 2 (45) 1 (60) 1 (8) 1 (11) 1 (4) 2 (11) 1 (1) 1 (15) 1 (6)

Body temperature greater than or equal to 388C. Liver transaminase enzymes at 1.5 times normal levels (concentrations of aspartate transaminase of greater than or equal to 38 IU/liter and of alanine transaminase of greater than or equal to 44 IU/liter). c Dyspnea, cough, expectoration, hemoptysis, or chest pain, associated with chest radiographic abnormalities (when radiographs were available). d The beginning of clinical signs was defined as day 0. e 1, positive culture; 2, negative culture. b

Downloaded from http://jcm.asm.org/ on December 14, 2017 by guest

patients with acute Q fever. Sixty-six blood samples from 66 patients were collected prior to any specific antibiotic treatment. From those samples, 11 (17%) blood cultures positive for C. burnetii were obtained. Cultures of the 160 blood samples collected after the initiation of specific antibiotic therapy remained negative. Clinical, serological, and therapeutic details for 10 of the 11 culture-positive patients are presented in Table 2 (for one patient, clinical data were not available). All patients were males. None of them was hospitalized for suspicion of acute Q fever. Patient 4A had received tetracyclines, but blood was collected 25 days after the antibiotics had been stopped. Three patients suspected of malaria were hospitalized after they developed a febrile illness after recent foreign travel (French Guiana for patient 2A, Senegal for patient 3A, and Cambodia for patient 7A). For seven patients, serology was negative when C. burnetii was isolated. Patient 6A underwent laparotomy because of suspected acute cholecystis. For the 10 patients who were reported on, 9 had fever, 6 had hepatitis, 3 had pneumonia, and 2 had all three symptoms. Chronic Q fever. We performed 162 blood cultures for the 71 patients with chronic Q fever. Seventeen samples from 17 patients were collected prior to any specific antibiotic treatment. Of those samples, 9 (53%) blood cultures positive for C. burnetii were obtained. Cultures of the 145 blood samples

collected after the initiation of specific antibiotic therapy remained negative. Clinical, serological, therapeutic, and blood and cardiac valve culture data are presented in Table 3. All serological results indicated evidence of chronic Q fever when C. burnetii was isolated. For patients 2C, 3C, and 6C, cultures derived from blood and valves were both positive. For patient 2C, blood and valve cultures were positive 1 year after the prescription of antibiotic therapy, because there was no compliance with the prescription. For six patients not reported on here, blood samples collected at the time of valvular replacement were culture negative, even though C. burnetii was isolated from the cardiac valves. Of the 20 strains isolated from blood, 10 were positive by immunofluorescence at day 6 of incubation and 10 were positive at day 14. All strains isolated from cardiac valves were detected by the sixth day of incubation. For patient 3C, blood collected during surgery was culture positive at day 14, whereas the cardiac valves were culture positive on day 6.

VOL. 33, 1995

BLOOD CULTURE FOR DIAGNOSIS OF Q FEVER

3131

TABLE 3. Clinical, serological, therapeutic, and blood and valve culture results for chronic Q-fever patients with blood cultures positive for C. burnetii Anti-phase I IgG antibody titer (mo/ day/yr)

Clinical presentation

1C

Endocarditis

Doxycycline and quinolones (5/25/92)

2C

Endocarditis

Doxycycline and quinolones (3/15/92)a

3C

Endocarditis

Doxycycline and hydroxy chloroquine (9/8/93) Doxycycline and hydroxy chloroquine (3/20/92)

4C

Endocarditis

Doxycycline and hydroxy chloroquine (8/11/92)

5C 6C

Endocarditis Endocarditis

Doxycycline and hydroxy chloroquine (10/18/91) Doxycycline, quinolones, and hydroxy chloroquine (4/15/91)

7C

Endocarditis

Doxycycline, quinolones, and hydroxy chloroquine (3/26/91)

8C 9C

Endocarditis Endocarditis

Doxycycline and hydroxy chloroquine (3/15/91)

a

Treatment and date of initiation (mo/day/yr)

Blood culture (mo/day/yr)

Cardiac valve culture (mo/day/yr)

12,800 (5/22/92) 12,800 (9/8/92) 1,600 (3/10/92) 1,600 (5/12/93)

1 (5/22/92) 2 (9/8/92)

1 (6/24/92)

1 (5/12/93)

1 (5/12/93)

12,800 (2/4/92) 6,400 (3/31/92) 51,200 (8/11/92) 200 (9/14/93) 1,600 (10/6/91) 1,600 (4/8/91) 3,200 (2/14/92) 409,600 (3/21/91) 51,200 (5/22/91) 204,800 (8/18/91) 1,600 (1/25/91) 1,600 (3/2/91) 1,600 (9/20/91)

1 (2/4/92) 2 (3/31/92) 1 (8/11/92) 2 (9/14/93) 1 (10/6/91) 1 (4/8/91)

1 (2/4/92)

1 (3/21/91) 2 (5/22/91)

1 (8/18/91)

1 (1/25/91) 1 (3/2/91) 2 (9/20/91)

1 (2/14/91)

1 (4/8/91)

This patient was not compliant with this treatment.

Seven culture-positive acute Q-fever patients had no detectable specific antibodies when C. burnetii was isolated. This observation serves to remind physicians that such a serological result does not eliminate Q fever as a possible diagnosis, and a convalescent-phase serum sample taken 3 weeks later is necessary. Furthermore, these observations highlight the usefulness of culture results in the primary diagnosis of acute Q fever. Although cultures may take up to 14 days, a positive result can still be obtained prior to serological evidence of infection. Five of the patients with acute cytolitic hepatitis were tested for viral hepatitis A, B, and C and cytomegalovirus but were negative. Blood samples for the cultivation of C. burnetii are particularly useful for etiological diagnosis of acute hepatitis; however, unlike the case with viral hepatitis, serology may be negative at the time of cytolysis. A sample was taken from patient 10A after complete recovery. Patient 4A received tetracycline antibiotic treatment, but after discontinuation of treatment, this patient had a positive blood culture despite having made a complete recovery. These cases demonstrate that a positive C. burnetii blood culture is possible for asymptomatic patients and show that although short-duration treatment with an appropriate antibiotic is clinically effective for acute Q fever, it does not eradicate C. burnetii and therefore does not prevent possible evolution towards chronic disease. Thus, with acute Q fever, cultures should be performed even for afebrile patients. In a serosurvey of 942 blood donors conducted in the area of Marseille, we found a Q-fever seroprevalence of 4.03 per 100 inhabitants (12). The high prevalence of Q fever in our region together with the possibility of asymptomatic cases perhaps suggest that blood transfusion is an alternative mode of Qfever transmission. Such a possibility may also explain the higher prevalence of Q fever observed for human immunodeficiency virus-seropositive intravenous drug abusers (10) and dialysis patients (5). Fifty-three percent of untreated chronic Q-fever patients yielded positive blood cultures. As with the acute Q-fever patients, only one sample was collected during the bacteremic phase. However, unlike the case with the acute disease, blood

culture and serology were positive at the same time. Because the clinical evolution of chronic Q fever is slow, serology is always positive when the diagnosis is performed. Isolation of the bacterium therefore allows for confirmation of the serological diagnosis, but it is not useful for primary diagnosis. In patients with Q-fever endocarditis, the percentage of isolation is 53%, compared with an isolation rate of 95% for infective endocarditis caused by more usual agents (20). For six patients with Q-fever endocarditis not described here, blood samples collected at the time of valvular replacement were negative, even though C. burnetii was isolated from the cardiac valves. Cell cultures inoculated with valvular material yield detectable organisms sooner than those inoculated with blood samples. Whereas all cardiac valve-seeded cultures were direct immunofluorescence positive by the sixth day of incubation, 50% of blood-seeded cultures yielded no detectable organisms until testing on day 14. A lower C. burnetii blood inoculum may be responsible for this lower sensitivity. Blood culture is useful in monitoring the success of therapy for Q fever, because cultures should remain negative following the initiation of treatment. Positive cultures during treatment may indicate that the patient is failing to follow the prescribed regimen correctly (as was the case with patient 2C). The C. burnetii isolation method used in this study is suitable for all clinical laboratories with cell culture facilities and biohazard facilities. Physicians should be asked to collect samples prior to antibiotic treatment. For acute Q fever, blood cultures should be used for diagnosis at the onset of the illness, because early serology is insensitive. For chronic Q fever, isolation of C. burnetii confirms the etiological diagnosis of blood culturenegative endocarditis and is useful for therapeutic follow-up. ACKNOWLEDGMENT We thank T. J. Marrie for helpful review of the manuscript. REFERENCES 1. Brouqui, P., H. Tissot Dupont, M. Drancourt, Y. Berland, J. Etienne, C. Leport, F. Golstein, P. Massip, M. Micoud, A. Bertrand, and D. Raoult. 1993. Chronic Q fever. Ninety-two cases from France, including 27 cases without endocarditis. Arch. Intern. Med. 153:642–648.

Downloaded from http://jcm.asm.org/ on December 14, 2017 by guest

Patient

3132

MUSSO AND RAOULT

isolates of Coxiella burnetii. Antimicrob. Agents Chemother. 35:2070–2077. 14. Raoult, D., G. Vestris, and M. Enea. 1990. Isolation of 16 strains of Coxiella burnetii from patients by using a sensitive centrifugation cell culture system and establishment of the strains in HEL cells. J. Clin. Microbiol. 28:2482– 2484. 15. Ripa, K. T., and P.-A. Mårdh. 1977. Cultivation of Chlamydia trachomatis in cycloheximide-treated McCoy cells. J. Clin. Microbiol. 6:328–331. 16. Shuster, E. A., J. S. Beneke, G. E. Tegtmeier, G. R. Pearson, C. A. Gleaves, A. D. Wold, and T. F. Smith. 1985. Monoclonal antibody for rapid laboratory detection of cytomegalovirus infections: characterization and diagnostic application. Mayo Clin. Proc. 60:577–585. 17. Smadel, J. E., and E. B. Jackson. 1964. Rickettsial infections, p. 743–773. In E. H. Lennette and N. J. Schmidt (ed.), Diagnostic procedure for viral and rickettsial diseases. American Public Health Association, New York. 18. Tissot Dupont, H., D. Raoult, P. Brouqui, F. Janbon, D. Peyramond, P. J. Weiller, C. Chicheportiche, M. Nezri, and R. Poirier. 1992. Epidemiologic features and clinical presentation of acute Q fever in hospitalized patients: 323 French cases. Am. J. Med. 93:427–434. 19. Tissot Dupont, H., X. Thirion, and D. Raoult. 1994. Q fever serology: cutoff determination for microimmunofluorescence. Clin. Diagn. Lab. Immunol. 1:189–196. 20. Werner, A. S., C. G. Cobbs, and D. Kaye. 1967. Studies of the bacteremia of bacterial endocarditis. JAMA 202:199. 21. Williams, J. C., L. A. Thomas, and M. G. Peacock. 1986. Humoral immune response to Q fever: enzyme-linked immunosorbent assay antibody response to Coxiella burnetii in experimentally infected guinea pigs. J. Clin. Microbiol. 24:935–939. 22. Yeaman, M. R., L. A. Mitscher, and O. G. Baca. 1987. In vitro susceptibility of Coxiella burnetii to antibiotics, including several quinolones. Antimicrob. Agents Chemother. 31:1079–1084. 23. Yu, X., and D. Raoult. 1994. Serotyping Coxiella burnetii isolates from acute and chronic Q fever patients by using monoclonal antibodies. FEMS Microbiol. Lett. 117:15–20.

Downloaded from http://jcm.asm.org/ on December 14, 2017 by guest

2. Donowitz, G. R., and G. L. Mandell. 1995. Acute pneumonia, p. 619–637. In G. L. Mandell, J. E. Bennett, and R. Dolin (ed.), Principles and practice of infectious diseases, 4th ed. Churchill Livingstone, Ltd., Edinburgh. 3. Dupuis, G., O. Peter, D. Pedroni, and J. Petite. 1985. Aspects cliniques observe´s lors d’une ´epide ´mie de 415 cas de fie `vre Q. Schweiz. Med. Wochenschr. 115:814–818. 4. Fiset, P., and T. E. Woodward. 1982. Q fever, p. 435–448. In A. S. Evans and H. A. Feldam (ed.), Bacterial infections of humans. Plenum Medical Book Co., New York. 5. Leonetti, F., D. Raoult, B. Dussol, B. Brunet, and Y. Berland. 1994. Chronic Q fever in hemodialysis patients. Nephron 67:231–233. 6. Marrero, M., and D. Raoult. 1989. Centrifugation-shell vial technique for rapid detection of Mediterranean spotted fever rickettsia in blood culture. Am. J. Trop. Med. Hyg. 40:197–199. 7. Marrie, T. J. 1985. Pneumonia and meningo-encephalitis due to Coxiella burnetii. J. Infect. 11:59–61. 8. Micoud, M., J. P. Brion, J. C. Boulard, J. L. Magne, B. Gratacap, J. P. Stahl, I. Farah, and D. Raoult. 1991. Infection of aortic aneurysm with Coxiella burnetii. Lancet 338:584. 9. Peter, O., G. Dupuis, W. Burgdorfer, and M. Peacock. 1985. Evaluation of the complement fixation and indirect immunofluorescence test in the early diagnosis of primary Q fever. Eur. J. Clin. Microbiol. 4:394–396. 10. Raoult, D., P. Y. Levy, H. Tissot Dupont, C. Chicheportiche, C. Tamalet, J. A. Gastaut, and J. Salducci. 1993. Q fever and HIV infection. AIDS 7:81–86. 11. Raoult, D., and A. Stein. 1994. Q fever during pregnancy—a risk for women, fetuses, and obstetricians. N. Engl. J. Med. 330:371. (Letter.) 12. Raoult, D., B. Toga, H. Chaudet, and C. Chiche-Portiche. 1987. Rickettsial antibody in southern France: antibodies to Rickettsia conorii and Coxiella burnetii among urban, suburban and semi-rural blood donors. Trans. R. Soc. Trop. Med. Hyg. 81:80–81. 13. Raoult, D., H. Torres, and M. Drancourt. 1991. Shell-vial assay: evaluation of a new technique for determining antibiotic susceptibility, tested in 13

J. CLIN. MICROBIOL.