In patients with cystic fibrosis, overwhelming infection caused by Burkholderia cepacia is a recognized but infrequent clinical entity. Although the "cepacia syndrome" is described in the current literature, its pathogenesis is not well understood. We describe an example of this clinical phenomenon in an adolescent patient with cystic fibrosis in whom fingerprints of B. cepacia suggest that she may have carried the organism for an extended period before becoming ill with the cepacia syndrome.
*Eastern Virginia Medical School, and †Children's Hospital of the King's Daughters, Norfolk, VA and ‡Department of Pediatrics, School of Medicine, Wake Forest University, Winston-Salem, NC.
Address correspondence and reprint requests to Karl Karson, MD, Wake Forest University Baptist Medical Center, Medical Center Boulevard, Winston-Salem, NC 27157. Email: email@example.com.
The patient was a 15-year-old white female with cystic fibrosis (CF) that was diagnosed as a neonate with meconium ileus. For the first 8 years of her life, she was followed at another CF center in the mid Atlantic region. She had no prior documented microbiological records of Burkholderia cepacia at that center. She was diagnosed with allergic bronchopulmonary aspergillosis. However, she was unable to be weaned from daily steroids because of increased immunoglobulin E levels and increased symptoms. Her most recent immunoglobulin E level was 152 kU/L 4 days before her last hospital admission but was as high as 814 kU/L in the past. She also had recurrent sinonasal polyposis and sinusitis. Her past hospital admissions were for functional endoscopic sinus surgery and nasal septoplasty at age 14. She did require intravenous antibiotics for these hospital courses. Additionally, she was taking cephalexin daily. She was evaluated regularly as an outpatient. She did not have any weight loss (5% weight, 3% height). Her pulmonary function tests (PFTs) were forced vital capacity (FVC) of 105% of predicted value, forced expiratory volume in the first second (FEV1) of 76% of predicted value, and forced expiratory flow (FEF) of 50% to 44% of predicted value when she was 10 years old; 5 months before her last admission, her PFTs were FVC of 91%, FEV1 of 68%, and FEF of 50% to 43%. Her chest radiographs consistently demonstrated focal areas of bronchiectasis, although at the age of 9 years, she did have a focal infiltrate and crackles on examination. She was treated with oral antibiotics, and the infiltrate was resolved. She actively participated in field sports in high school.
Sputum cultures were obtained at regular intervals for surveillance. Pseudomonas aeruginosa was first isolated in 1990 and was only infrequently cultured from these, whereas α-hemolytic streptococci, Neisseria species, and nontypeable Haemophilus influenzae were prominent. Serratia marcescens, Staphylococcus aureus, and Aspergillus species were frequently isolated but always "rare" or "few" in number. A culture obtained in 1999 contained a gram-negative rod that could not be grown for identification or susceptibility testing. A culture obtained in November 2002 contained moderate B. cepacia (genomovar III) as well as α-streptococci, Neisseria species, and rare Aspergillus species. This did not clinically impact her lung disease, her PFTs, or her activity. Seven subsequent sputum cultures obtained between December 2002 and March 2004 did not grow B. cepacia but continued to show her other flora plus P. aeruginosa on occasion.
Two weeks before her final admission, she developed a new nonproductive cough. She was placed on clarithromycin, but she did not improve. Her cough became productive but without hemoptysis. She was placed on a tapering course of oral steroids for presumed allergic bronchopulmonary aspergillosis. Her PFT had been stable 2 weeks before admission (FVC, 98% of predicted value; FEV1, 76% of predicted value; and FEF, 50%-50%) but acutely worsened on the day of her last admission (FVC, 88% of predicted value; FEV1, 65% of predicted value; FEF, 50%-41% of predicted value). Also, a chest radiograph obtained the same day revealed a new 1.5 × 2-cm round opacity in both lung fields. A chest computed tomographic scan (Fig. 1) was done and showed multiple 1- to 5-cm bilateral pulmonary nodules with central cavitation distributed throughout both lung fields. The differential diagnosis at that time included invasive pulmonary aspergillosis, invasive Pseudomonas infection and pulmonary nocardiosis.
At admission, she was intermittently febrile and had pleuritic chest pain that had started 48 hours earlier. She continued to have a productive cough. Her complete blood cell count demonstrated a leukocytosis that, in addition to her fever (temperature, 38°C-39.7°C), persisted throughout her hospital course. Her initial regimen included intravenous amphotericin B, ticarcillin/clavulanate (TIC/CLAV), intravenous tobramycin, and intravenous corticosteroids. Sputum cultures were obtained and subsequently grew α-hemolytic streptococci, Neisseria species, methicillin-susceptible S. aureus, nontypeable H. influenzae, and S. marescens. Admission blood cultures were sterile. Over the next 4 days, she developed an oxygen requirement (2L, nasal cannula). Voriconazole was substituted for amphotericin B, and trimethoprim/sulfamethoxazole (TMP/SMX) was started. A repeat sputum culture obtained 4 days after admission grew α-streptococci, Neisseria species, and rare Aspergillus species. A thorascopic lung biopsy was performed, and a nodule from the left lower lobe was excised. The histology showed massive consolidation with numerous microabscesses and foci of necrosis without evidence of invasive fungi. Cultures were placed on multiple media, including Burkholderia-selective medium oxidation-fermentation polymyxin bacitracin lactose agar. Intravenous tobramycin, TIC/CLAV, and TMP/SMX were continued postoperatively. An additional oxidase-positive, glucose-negative, and gram-negative rod was isolated from her sputum culture that subsequently grew poorly on subculture and yielded poor probability speciation results on biochemical testing and conflicting antibiotic susceptibility testing results by both broth dilution and Kirby-Bauer methods. At this time, her antibiotic therapy was switched from tobramycin and TIC/CLAV to meropenem and amikacin with transient clinical stabilization. The sputum isolate was presumptively identified 4 days later as B. cepacia (genomovar III). Her antibiotic regimen was again switched to ceftazidime, meropenem, tobramycin, TMP/SMX, and voriconazole.
After 2 days, the lung biopsy specimen grew an oxidase-positive, glucose-negative, and gram-negative rod with morphological characteristics similar to the sputum isolate. Again, biochemical testing gave poor probability results for speciation, and there was no growth on the oxidation-fermentation polymyxin bacitracin lactose agar medium. Isolates from the lung biopsy and sputum were sent to the Cystic Fibrosis Foundation Burkholderia cepacia Research Laboratory and Repository and were identified within 2 days as B. cepacia (genomovar III) by random amplification of polymorphic DNA (RAPD) typing. The patient showed neither improvement nor stabilization in her clinical status. At her ninth day of hospitalization, she had a slow but progressive increase in both her oxygen requirement and her work of breathing. Her chest radiograph showed little change in the patchy bilateral opacities except for some central cavitation in the opacities. Her lung biopsy isolate was sent to the CF Referral Center for Susceptibility and Synergy Studies for testing. Two weeks after admission, she was transferred to the pediatric intensive care unit where a 48-hour trial of BiPaP for respiratory support was begun. After having failed her trial of BiPaP for 72 hours, she was electively intubated. After transient stabilization, she showed progressive hypercapnia over the ensuing 3 days, despite application of multiple ventilator protocols and positional changes, ultimately developing refractory bradycardia and hypotension that resulted in her death 3 days later. Blood cultures obtained 17 and 19 days after her admission grew B. cepacia (genomovar III) that was resistant to all antibiotics. Three days after her death, the antibiotic in vitro synergy testing results returned, showing a highly resistant organism that was susceptible only to minocycline with additive effects with the combinations of doxycycline + meropenem, doxycycline + chloramphenicol, and chloramphenicol + meropenem. All her isolates were identified as B. cepacia (genomovar III) by RAPD, which did not match any fingerprints (by RAPD) from patients in our center but did match organisms (by RAPD) cultured from her previous CF center.
One of the first published descriptions of "cepacia syndrome" was in a case series from Isles et al1 published 23 years ago in which there was an increasing prevalence of patients with B. cepacia infection. These patients were found to have impairment of pulmonary function and a syndrome characterized by high fever, severe progressive respiratory failure, leukocytosis, and increased erythrocyte sedimentation rate. The cepacia strains were highly resistant to antibiotics and had a high case fatality rate.
Another recent published case report by Blackburn et al2 describes an adolescent male patient who was chronically colonized with B. cepacia for 9 years, then clinically deteriorated. His final hospital course was also characterized by fever, leukocytosis, and increased C-reactive protein. He also had bilateral patchy infiltrates on chest radiograph and positive sputum and blood cultures for Burkholderia multivorans (genomovar II).
Our patient shared similar features with both of these published reports. For most of her life, she had mild-to-moderate disease and then progressive pulmonary decline with cavitary pneumonia, high fever, leukocytosis, bacteremia, sepsis, pulmonary microabscess formation, and ultimately death. Unlike our other patients with B. cepacia, her B. cepacia fingerprint was not found in our center. Therefore, she may have carried this organism before her move, or she may have been transiently colonized or may have acquired it more recently during a visit with friends from her first center. From the B. cepacia isolates that we were able to retrieve, we know that she was at least intermittently colonized. She is of interest because of the time lapse between sputum cultures that were positive for B. cepacia. Her first retrieved B. cepacia (genomovar III) isolate was 17 months before this hospitalization. Additionally, there was a gram-negative rod obtained 5 years prior from a sputum sample that could not be identified. Her subsequent sputum cultures may not have grown secondary to inadequate production and/or collection, despite intensive efforts from experienced laboratory technicians. However, other CF patients with B. cepacia at our center do not have B. cepacia (genomovar III), whereas her B. cepacia did match those of B. cepacia (genomovar III) cultured from her previous CF center.
Courtney et al3 looked at clinical outcomes of B. cepacia complex in adults with CF who were infected with B. cepacia genomovar II or III for at least 6 years and compared them with CF patients with chronic P. aeruginosa infection. Patients with B. cepacia (genomovar III) had a significantly greater rate of FEV1 decline compared with those infected with B. multivorans (genomovar II) or P. aeruginosa. Also, within the B. cepacia (genomovar III) group, 7 of the 16 patients died. Of those that died, half had cepacia syndrome. Currently, critical facts are still unknown about when B. cepacia is initially acquired and how long a CF patient may be colonized before developing cepacia syndrome.
The retrieval, identification, and classification of B. cepacia is evolving and allowing for a more discriminating classification.4 In the meantime, for the physician, consider cepacia syndrome in the CF patient who develops a progressive and precipitous pulmonary decline. Recognize that B. cepacia is difficult to identify by standard microbiological techniques available in most hospitals; current guidelines recommend sending CF isolates of Burkholderia species to the Cystic Fibrosis Foundation Burkholderia cepacia Research Laboratory and Repository for identification, speciation, and molecular typing.5
The authors thank Dr J. LiPuma for analysis of all the B. cepacia isolates.
1. Isles A, Maclusky I, Corey M, et al. Pseudomonas cepacia
infection in cystic fibrosis: an emerging problem. J Pediatr
2. Blackburn L, Brownlee K, Conway S, et al. "Cepacia syndrome" with Burkholderia multivorans
, 9 years after initial colonization. J Cyst Fibros
3. Courtney J, Dunbar K, McDowell A, et al. Clinical outcome of Burkholderia cepacia
complex infection in cystic fibrosis adults. J Cyst Fibros
4. LiPuma J. Update on the Burkholderia cepacia
complex. Curr Opin Pulm Med
5. Saiman L, Siegel J, Cystic Fibrosis Foundation. Infection control recommendations for patients with cystic fibrosis: microbiology, important pathogens, and infection control practices to prevent patient-to-patient transmission. Infect Control Hosp Epidemiol