Literature DB >> 35146052

The First Reported Case of Bordetella pertussis Bacteremia in a Patient With Human Immunodeficiency Virus Infection.

Yuji Wakimoto1, Nao Otsuka2, Yasuaki Yanagawa1, Kentaro Koide2, Kazunari Kamachi2, Keigo Shibayama2, Katsuji Teruya1, Yoshimi Kikuchi1, Shinichi Oka1.   

Abstract

We describe a case of bacteremia in a human immunodeficiency virus-infected patient caused by a Bordetella pertussis strain lacking 2 major virulence factors, filamentous hemagglutinin and fimbriae. Although B pertussis bacteremia is uncommon, physicians should be aware that even attenuated B pertussis strains can cause invasive infection in immunocompromised patients. Bordetella pertussis is a gram-negative coccobacillus that causes a severe paroxysmal coughing disease known as whooping cough or pertussis. Bordetella pertussis colonizes the epithelial cells of the human respiratory tract, and the organisms are typically isolated from nasopharynx. We describe a case of B pertussis bacteremia in a patient with human immunodeficiency virus (HIV) infection. Interestingly, the isolate recovered from blood culture did not produce the major virulence factors, filamentous hemagglutinin (FHA) and fimbriae (FIM). Previously, 3 cases of B pertussis bacteremia were reported in the literature. We discuss the features of B pertussis bacteremia.
© The Author(s) 2022. Published by Oxford University Press on behalf of Infectious Diseases Society of America.

Entities:  

Keywords:  Bordetella pertussis; HIV; antigen deficiency; bacteremia

Year:  2022        PMID: 35146052      PMCID: PMC8825563          DOI: 10.1093/ofid/ofac020

Source DB:  PubMed          Journal:  Open Forum Infect Dis        ISSN: 2328-8957            Impact factor:   3.835


CASE REPORT

A 49-year-old man with HIV infection visited our emergency department in October 2019 complaining of intermittent fever, chest pain, severe productive cough, and dyspnea. He had been diagnosed with HIV infection 14 years previously; however, 1 year before admission to our hospital, he discontinued antiretroviral therapy. His vaccination history was unknown. During physical examination, the patient was alert and fully oriented. His body temperature was 36.7°C, blood pressure was 86/54 mm Hg, pulse rate was 180 beats per minute with atrial fibrillation, respiratory rate was 20 breaths per minute, and oxygen saturation was 98% on room air. Blood tests revealed leukocytosis (white blood cell count, 69 910/µL), a strong inflammatory response (C-reactive protein level, 40.69 mg/dL), acute renal failure (serum creatinine level, 7.81 mg/dL), and metabolic acidosis (pH 7.27 and HCO3- 14.2 mEq/L). His CD4 count was 134 cells/µL and HIV viral load was 108 000 copies/mL. Chest auscultation revealed no respiratory sounds in the upper left lung field. Chest computed tomography revealed consolidation in the left upper lobe with air bronchograms. Histological findings from lung biopsy samples revealed that macrophages were mainly found in the alveoli and fibrillization had occurred along the alveolar duct. No pathological organisms were observed by Grocott methenamine silver stain or Gram stain. Analysis of paired sera (day 7 and day 43 of hospitalization) showed that levels of his immunoglobulin G antibodies against pertussis toxin (anti-PT IgG) and filamentous hemagglutinin (anti-FHA IgG) were significantly elevated during 5 weeks of hospitalization. He was initially treated for suspected bronchopneumonia using ceftriaxone for 5 days and levofloxacin for 16 days. After a 16-day course of antibiotics, his fever and respiratory symptoms did not respond to therapy and no improvement was observed on chest radiography. At that time, we remained unsure whether the sustained radiological findings had an active infectious etiology because polymerase chain reaction (PCR) results from lung biopsy samples were still pending. Moreover, because antimicrobial susceptibility test results for the causative B pertussis isolate (see below) were unavailable, we administered azithromycin for 14 days for definitive therapy of B pertussis infection. Following antimicrobial therapy, pulmonary sequelae were radiologically detected on computed tomography and he was diagnosed with secondary organizing pneumonia (Figure 1). Combination antiretroviral therapy with raltegravir, tenofovir alafenamide, and emtricitabine was administered starting on day 17. By day 39, his respiratory symptoms gradually improved. His CD4 count reached 511 cells/µL and HIV RNA was undetectable by day 50. The patient was discharged on day 65. During postdischarge follow-up, no relapse of pulmonary disorder including immune reconstitution syndrome was observed.
Figure 1.

Chest computed tomographic (CT) images of a human immunodeficiency virus (HIV)–infected patient with Bordetella pertussis bacteremia. A, The chest CT scan showed pulmonary consolidation with air bronchogram in the left upper lobe. B, Even after the treatment with azithromycin for 14 days following levofloxacin for 16 days, the lesion was not improved.

Chest computed tomographic (CT) images of a human immunodeficiency virus (HIV)–infected patient with Bordetella pertussis bacteremia. A, The chest CT scan showed pulmonary consolidation with air bronchogram in the left upper lobe. B, Even after the treatment with azithromycin for 14 days following levofloxacin for 16 days, the lesion was not improved. Haemophilus influenzae was found in a sputum taken at the time of admission, and small gram-negative bacilli were detected from blood cultures started on the day of admission (day 0) after 6 days of incubation. Using matrix-assisted laser desorption/ionization–time of flight mass spectrometry (MALDI Biotyper System, Bruker Daltonics, Germany) in combination with PCR analyses [1, 2], the organism was identified as B pertussis and the isolate was designated as BP611. Antimicrobial susceptibility of the isolate was examined using the Etest technique (bioMérieux Japan). The isolate was susceptible to all tested antibiotics (ampicillin, ceftriaxone, norfloxacin, erythromycin, gentamicin, and tetracycline). Interestingly, the isolate did not produce major B pertussis virulence factors including FHA and FIM (Fim2 and Fim3) (Figure 2).
Figure 2.

Protein analysis and genetic mechanisms of filamentous hemagglutinin (FHA) and fimbriae (FIM) deficiency in Bordetella pertussis BP611. A, Coomassie brilliant blue staining of B pertussis total protein. Five micrograms of protein per lane was run on 12.5% polyacrylamide gel. M: molecular marker. Lane 1: B pertussis 18323; lane 2: Tohama I; lane 3: BP611. B, Immunoblot analysis. Rabbit anti-Fim2 immunoglobulin G (IgG) (number CSB-PA356626LA01BUA) and anti-Fim3 IgG (number CSB-PA322971LA01BUA) antibodies were purchased from Cusabio, China. Anti-Fim3 IgG reacts to both Fim2 and Fim3. Upper and lower arrows indicate Fim2 (22.5 kDa) and Fim3 (22.0 kDa), respectively. Production of Fim2 and Fim3 was also confirmed by enzyme-linked immunosorbent assay method by using monoclonal antibodies against Fim2 and Fim3 (National Institute for Biological Standards and Control code numbers 04/154 and 04/156, respectively). C, Genetic mechanisms for FHA and FIM deficiencies. The sequence alignments of fhaB, promoter regions of fim2 (Pfim2) and fim3 (Pfim3), between B pertussis Tohama I and BP611 are shown. In Pfim2 and Pfim3, the transcriptional start site was counted as +1, which was previously reported [3]. Antigen production patterns for each strain are as follows: B pertussis 18323 (FHA+/Fim2+/Fim3+), Tohama I (FHA+/Fim2+/Fim3–), and BP611 (FHA–/Fim2–/Fim3–). The sequence reads of BP611 are available under BioProject number PRJDB9628 and DNA Data Bank of Japan Sequence Read Archive accession number DRR220690.

Protein analysis and genetic mechanisms of filamentous hemagglutinin (FHA) and fimbriae (FIM) deficiency in Bordetella pertussis BP611. A, Coomassie brilliant blue staining of B pertussis total protein. Five micrograms of protein per lane was run on 12.5% polyacrylamide gel. M: molecular marker. Lane 1: B pertussis 18323; lane 2: Tohama I; lane 3: BP611. B, Immunoblot analysis. Rabbit anti-Fim2 immunoglobulin G (IgG) (number CSB-PA356626LA01BUA) and anti-Fim3 IgG (number CSB-PA322971LA01BUA) antibodies were purchased from Cusabio, China. Anti-Fim3 IgG reacts to both Fim2 and Fim3. Upper and lower arrows indicate Fim2 (22.5 kDa) and Fim3 (22.0 kDa), respectively. Production of Fim2 and Fim3 was also confirmed by enzyme-linked immunosorbent assay method by using monoclonal antibodies against Fim2 and Fim3 (National Institute for Biological Standards and Control code numbers 04/154 and 04/156, respectively). C, Genetic mechanisms for FHA and FIM deficiencies. The sequence alignments of fhaB, promoter regions of fim2 (Pfim2) and fim3 (Pfim3), between B pertussis Tohama I and BP611 are shown. In Pfim2 and Pfim3, the transcriptional start site was counted as +1, which was previously reported [3]. Antigen production patterns for each strain are as follows: B pertussis 18323 (FHA+/Fim2+/Fim3+), Tohama I (FHA+/Fim2+/Fim3–), and BP611 (FHA–/Fim2–/Fim3–). The sequence reads of BP611 are available under BioProject number PRJDB9628 and DNA Data Bank of Japan Sequence Read Archive accession number DRR220690.

DISCUSSION

Bordetella pertussis rarely causes blood infections; this is only the fourth report of B pertussis bacteremia. The general features of B pertussis bacteremia are summarized in Table 1 [4, 5, 8]. All patients were adults (mean age, 56 years) and had underlying immunocompromised disease prior to B pertussis infection. Fever or intermittent fever was commonly observed.
Table 1.

Features of 4 Patients With Bordetella pertussis Bacteremia

Factor Case and Year of Publication [Reference]
Patient 1 Patient 2 Patient 3Patient 4
1994 [4]2004 [5]2006 [6]Present Case
Age, sex31, male82, female63, male49, male
Underlying diseaseGranulomatosis with polyangiitisMultiple myelomaMultiple myelomaHIV infection
SymptomNonproductive coughIntermittent fever with chillsProgressive shortness of breathWheezingPneumoniaCoughFeverLung ralesDyspneaPneumonia1st admission:Productive coughFeverBronchial ralesPneumonia 2nd admission:Dry coughHoarsenessFeverDiffuse bilateral wheezing Productive coughIntermittent feverDyspneaChest painPneumonia
Temperature36.8ºC39ºC39.7ºC38.8ºC36.7ºC
WBC count23 200/µLNANormal12 700/µL69 910/µL
CRPNANA13.1 mg/dL27.0 mg/dL40.69 mg/dL
SerologyNANANAAnti-PT + FHA IgA, highAnti-PT + FHA IgG, highAnti–B pertussis whole cell, negativeDay 7: Anti-PT IgG, 79 EU/mLaAnti-FHA IgG, 9 EU/mLaDay 43: Anti-PT IgG, ≥160 EU/mLaAnti-FHA IgG, 55 EU/mLa
Chest radiographyRight lower lung infiltratesAbsence of cardiomegaly, a cavity bleb in the right middle lobeSutures from a previous surgery in the right upper lobeDiffuse left lung infiltratesNormalNormalLeft upper lung infiltrates, followed by fibrous organizationLeft lower lung nodular shadow
B pertussis–positive specimenBloodBloodBloodNasopharynxBlood
Time required for blood culture positive6 daysNA6 daysNA6 days
Blood infection therapyCeftizoxime, erythromycin, gentamicinMultiple nonmacrolide drugsCefotaximeCefotaxime, benzylpenicillin, clarithromycinCeftriaxone, levofloxacin, azithromycin
OutcomeDiedDiedCuredCured

Abbreviations: CRP, C-reactive protein; EU, enzyme-linked immunosorbent assay units; FHA, filamentous hemagglutinin; HIV, human immunodeficiency virus; IgA, immunoglobulin A; IgG, immunoglobulin G; NA, not available; PT, pertussis toxin; WBC, white blood cell.

The antibody titers can be calculated in international units (IU)/mL using the following unit conversions: anti-PT IgG, IU/mL = 1.19 × EU/mL; anti-FHA IgG, IU/mL = 1.01 × EU/mL [7].

Features of 4 Patients With Bordetella pertussis Bacteremia Abbreviations: CRP, C-reactive protein; EU, enzyme-linked immunosorbent assay units; FHA, filamentous hemagglutinin; HIV, human immunodeficiency virus; IgA, immunoglobulin A; IgG, immunoglobulin G; NA, not available; PT, pertussis toxin; WBC, white blood cell. The antibody titers can be calculated in international units (IU)/mL using the following unit conversions: anti-PT IgG, IU/mL = 1.19 × EU/mL; anti-FHA IgG, IU/mL = 1.01 × EU/mL [7]. All patients suffered from paroxysmal cough and presented clinical manifestations of pneumonia. However, because of the absence of direct laboratory findings, it was unclear whether B pertussis was the causative agent of respiratory symptoms in these patients and how the organism entered the bloodstream. Our patient developed organized pneumonia, but B pertussis was not detected by real-time PCR in bronchoalveolar lavage fluid (day 15) or lung biopsy samples (day 31), perhaps owing to the late timing of sampling following antimicrobial administration. Nasopharyngeal specimens were not obtained because B pertussis infection was initially not suspected. As for the potential portal of entry for B pertussis into the bloodstream, our patient had lung lesions, necrotizing ulcerative gingivitis, and erosions in the lower extremities. Given the low frequency of primary bloodstream infection and the typical pertussis symptoms of our patient, we speculate that B pertussis entered the bloodstream via lung lesions after inducing severe pneumonia. In general, most adolescents and adults with B pertussis infections present with milder signs and symptoms than infants and children. Especially in the early stages of infection (the catarrhal stage), clinical signs resemble those of other viral infections. Thus, the possibility of B pertussis infection is often overlooked unless physicians suspect this organism and perform B pertussis–specific nucleic acid amplification tests. Kilgore et al reported that adult pertussis often results in clinical complications including insomnia, apnea, weight loss, urinary incontinence, syncope, and rib fracture because of delayed diagnosis [6]. Pneumonia was a less common manifestation, occurring in only 0.6%–8.0% of adolescents and adults with B pertussis infection. However, pneumonia was observed in all 4 documented B pertussis bacteremia cases (Table 1). Therefore, pneumonia resulting from advanced pertussis may be a signal of bacteremia in immunocompromised patients following B pertussis infection. The slow growth of B pertussis from blood cultures impacts antimicrobial therapy. In the cases of patient 1, patient 3, and our patient (patient 4), blood cultures required 6 days of incubation before turning positive. The current standard for blood cultures performed using continuous-monitoring blood systems is 5 days. Bourbeau and Foltzer demonstrated that 97.4% of clinically significant isolates were recovered within the first 3 days of incubation and 93.8% were recovered within 2 days of incubation [9]. Our laboratory employs a 7-day incubation for routine blood cultures because of current regulatory requirements, and therefore B pertussis was detected on day 6 by chance. When treating bacteremia in immunocompromised patients, physicians should consider the possibility of B pertussis infection even if routine blood cultures are negative. When antimicrobial susceptibility was assessed by Etest, the B pertussis blood culture isolate was susceptible to β-lactams and fluoroquinolones. However, definitive therapy should be optimized using narrow-spectrum and target-specific drugs. Macrolides are first-line drugs for B pertussis infection, and the patient described here responded well to azithromycin. Recently, macrolide-resistant B pertussis strains have been emerging globally, especially in Asia [10, 11]. Therefore, antimicrobial susceptibility testing–based de-escalation is recommended for B pertussis infections. Although the B pertussis blood culture isolate described here was FHA deficient, our patient showed a significant increase in anti-FHA IgG titers during hospitalization. One possible explanation for this observation is the existence of another FHA-producing B pertussis strain; he could have been coinfected by a second FHA-producing B pertussis strain, or originally infected with an FHA-producing strain that subsequently lost FHA production following mutation in the fhaB gene. No FHA-producing strain was isolated from our patient, and, at present, no evidence is available to evaluate these hypotheses. An alternative explanation is antibody cross-reactivity. Nonencapsulated Haemophilus influenzae, Mycoplasma pneumoniae, and Chlamydia pneumoniae are known to produce FHA-like antigens, and their cross-reacting epitopes induce high titers of antibody to FHA [12-14]. Because H influenzae was isolated from the patient’s sputum, we speculate that the increase in his anti-FHA IgG titer may have been caused by its nonspecific antibody cross-reactivity. It should be noted that B pertussis causes systemic severe infections in immunocompromised patients even when the organism does not produce the major virulence factors, FHA and FIM. FHA and FIM are adhesion molecules that play cooperative roles in bacterial cell attachment to the respiratory epithelium [15]. They are also contained in acellular pertussis vaccines (aP) as protective antigens; aP contains several purified antigens—that is, PT, FHA, pertactin (PRN), and fimbriae (FIM: Fim2 and Fim3). Currently, PRN-deficient isolates are circulating in the aP-immunized population, and their prevalence has reached approximately 80% in some regions [16]. Although several studies have reported that patients infected with PRN-producing and PRN-deficient isolates show no major clinical differences, PRN- and PRN/FHA-deficient strains are suggested to have a selective advantage in evading aP immunity [16, 17]. As far as we know, this is the first report of an FHA/FIM-deficient B pertussis isolate, and therefore the clinical implications of the combined loss of FHA and FIM are still unknown, including whether loss of these factors alters clinical severity and/or diminishes vaccine immunity. Because antigen-deficient mutants may diminish vaccine effectiveness and represent public health threats, continued careful surveillance of B pertussis is necessary.

CONCLUSIONS

We report a case of B pertussis bacteremia in a patient with HIV infection. Bordetella pertussis blood infection is uncommon but can occur in immunocompromised patients who develop paroxysmal cough and pneumonia. Physicians should consider the possibility of B pertussis blood infection when treating immunocompromised patients. Continued surveillance of B pertussis is necessary to understand the clinical impact of antigen-deficient mutants.
  17 in total

1.  Unexpected isolation of Bordetella pertussis from a blood culture.

Authors:  W M Janda; E Santos; J Stevens; D Celig; L Terrile; P C Schreckenberger
Journal:  J Clin Microbiol       Date:  1994-11       Impact factor: 5.948

2.  Isolation of Bordetella pertussis in blood culture from a patient with multiple myeloma.

Authors:  Marius Trøseid; Tom Øystein Jonassen; Martin Steinbakk
Journal:  J Infect       Date:  2005-06-03       Impact factor: 6.072

3.  Novel architectural features of Bordetella pertussis fimbrial subunit promoters and their activation by the global virulence regulator BvgA.

Authors:  Qing Chen; Kimberly Baxter Decker; Philip E Boucher; Deborah Hinton; Scott Stibitz
Journal:  Mol Microbiol       Date:  2010-09       Impact factor: 3.501

4.  Cloning, expression, and DNA sequence analysis of genes encoding nontypeable Haemophilus influenzae high-molecular-weight surface-exposed proteins related to filamentous hemagglutinin of Bordetella pertussis.

Authors:  S J Barenkamp; E Leininger
Journal:  Infect Immun       Date:  1992-04       Impact factor: 3.441

Review 5.  Pertussis: Microbiology, Disease, Treatment, and Prevention.

Authors:  Paul E Kilgore; Abdulbaset M Salim; Marcus J Zervos; Heinz-Josef Schmitt
Journal:  Clin Microbiol Rev       Date:  2016-07       Impact factor: 26.132

6.  Cooperative roles for fimbria and filamentous hemagglutinin in Bordetella adherence and immune modulation.

Authors:  Erich V Scheller; Jeffrey A Melvin; Amanda J Sheets; Peggy A Cotter
Journal:  MBio       Date:  2015-05-26       Impact factor: 7.867

7.  A high seroprevalence of antibodies to pertussis toxin among Japanese adults: Qualitative and quantitative analyses.

Authors:  Takumi Moriuchi; Nao Otsuka; Yukihiro Hiramatsu; Keigo Shibayama; Kazunari Kamachi
Journal:  PLoS One       Date:  2017-07-10       Impact factor: 3.240

8.  Macrolide-Resistant Bordetella pertussis, Vietnam, 2016-2017.

Authors:  Kazunari Kamachi; Hong T Duong; Anh D Dang; T Hai; Do Do; Kentaro Koide; Nao Otsuka; Keigo Shibayama; Ha Thi Thu Hoang
Journal:  Emerg Infect Dis       Date:  2020-10       Impact factor: 6.883

9.  High Prevalence of Macrolide-Resistant Bordetella pertussis and ptxP1 Genotype, Mainland China, 2014-2016.

Authors:  Lijun Li; Jikui Deng; Xiang Ma; Kai Zhou; Qinghong Meng; Lin Yuan; Wei Shi; Qing Wang; Yue Li; Kaihu Yao
Journal:  Emerg Infect Dis       Date:  2019-12       Impact factor: 6.883

View more
  1 in total

1.  Invasive Bordetella pertussis Infection in Infants: A Case Report.

Authors:  Yang Liao; Wei-Ran Li; Yu Zhu; Shuang-Hong Luo; Qiong Liao; Chao-Min Wan
Journal:  Open Forum Infect Dis       Date:  2022-09-21       Impact factor: 4.423

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.