Identifying the Bacteria in Clinical Samples

Patient blood samples were sent to the Pennsylvania Department of Health (PADOH)’s Bureau of Laboratories for bacterial identification and to rule out Brucella species, according to standard Laboratory Response Network protocols. After negative Brucella results and the inability to identify the isolates, bacterial isolates were then sent to SBRL at CDC for bacterial identification. We identified the bacterial isolates recovered from each patient using previously described methods for16S rRNA gene sequence analysis and Clinical Laboratory and Standards Institute guidelines (1–3). We also used matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) analysis as a secondary identification method. We performed MALDI-TOF analysis on a Biotyper SMART (Bruker) by using an in-house library containing reference spectra from 6 strains of H. jordaniae, which is publicly searchable through MicrobeNet (https://microbenet.cdc.gov). All isolates yielded a MALDI-TOF score of >2.0 for a reliable species-level identification as H. jordaniae.

Medical Information

We conducted emergency department (ED) medical chart reviews for all 4 patients. Public health officials conducted interviews with patients 1, 2, and 3. Patient 4 could not be reached for interview.

Environmental Sample Collection

We identified 3 water locations in Pennsylvania for environmental sampling on the basis of patient interviews. Patient 2 identified locations A and B as their entry and exit points in the creek on the day of injury, near where patient 1 also reported injury. Patient 3 identified location C as the location where the patient entered the river on the day of injury, ≈50 miles from locations A and B.

We collected four 1-L water samples from locations A and B once a month throughout 2024 and four 1-L water samples from location C for 7 months in 2024. We collected water samples from the river’s surface, enabling the current to flow into the container without disturbing the river bottom sediment. At each sample collection site, we collected global positioning satellite coordinates, weather conditions, and water physiochemical measurements (pH, temperature, dissolved oxygen, conductivity, and specific conductance) by using a YSI Pro 2030 (YSI Incorporated).

Environmental Sample Processing and Testing

We shipped water samples to SBRL on cold packs on the day of sample collection. Because of resource constraints, a subset of the total number of collected samples were processed for DNA extraction. We filtered 1-L samples through MicroFunnel filter funnels containing a 0.22 μm filter (Cytiva), and we extracted genomic DNA from the filters by using the DNeasy PowerWater Kit (QIAGEN) according to manufacturer instructions. We analyzed extracted filter genomic DNA by using previously published methods for a real-time quantitative PCR (qPCR) specific to H. jordaniae (4). Modifications of methods for the screening of the submitted water samples included using the previously published qPCR with environmental samples and testing environmental genomic DNA extracted from water filters. We used 2 layers of controls, including positive (PCR amplification) and negative (nontemplate) controls for qPCR, as well as positive and negative filter extraction controls, which aided in confirming that targeted DNA could be collected through experimental filter extraction.

Ethics

This study did not meet the definition of human subjects research as defined in the US Code of Federal Regulations, Title 45 Part 46. Therefore, this study was not subject to review by an institutional review board.

Case Series

Patient 1

Patient 1 was a 28-year-old man who reported a scratch to his right distal leg in September 2020 (Figure 1). The injury occurred while participating in a recreational water activity in a south-central Pennsylvania creek. Within 24 hours of the injury, redness developed on the medial aspect of the patient’s right lower extremity. Two days later, swelling and pain developed at the location of the redness. The patient was initially treated by his primary care physician by telehealth appointment and prescribed cephalexin. After treatment was initiated, the redness started to improve but then became worse and spread down the patient’s leg, resulting in swelling of his ankle. The patient denied any fever or chills and was taking oral cephalexin (500 mg every 6 h for 10 days) as prescribed.

Ten days after the initial injury, the patient sought care at a local ED with a primary complaint of severe pain associated with the right lower extremity wound infection. Temperature was 37.1°C, heart rate 115 beats/min, respiratory rate 18 breaths/min, blood pressure 148/102 mm Hg, and oxygen saturation 99% on room air. Physical examination revealed swelling, tenderness, and erythema on the medial aspect of the right lower extremity from the ankle to halfway up the tibia. The patient’s remaining physical examination findings were unremarkable. Laboratory results at admission revealed an elevated leukocyte count of 21,800 cells/μL (reference range 3,900–9,500 cells/μL) (Table).

The patient was given 1 dose of vancomycin (1.25 g) in the emergency room before being hospitalized for sepsis and cellulitis and started on intravenous (IV) vancomycin (1.5 g every 8 h) and IV clindamycin (600 mg every 8 h) antimicrobial therapy. Radiographs and ultrasound of the right ankle showed a moderate amount of soft tissue swelling surrounding the ankle that was consistent with cellulitis. No fluid collection was observed in the ankle to indicate a septic joint, and no radiopaque foreign body or acute osseous abnormalities were observed. The patient clinically improved, was discharged after 3 days of in-hospital treatment, and was prescribed a 7-day course of oral clindamycin (300 mg every 6 h). A blood sample obtained at admission was cultured, and a bacterial isolate was recovered; 16S rRNA gene sequence analysis showed the isolate to be most closely related to H. jordaniae (reference strain no. DSM 28903T). A nearly full-length (1,406-bp) 16S rRNA gene sequence from the isolate shared 99.9% sequence similarity with DSM 28903T. On the basis of Clinical and Laboratory Standards Institute guidelines (3), we identified the isolate (GenBank accession no. PX060169) as H. jordaniae 3 months after the patient’s admission.

Patient 2

Patient 2 was a 34-year-old woman with a history of rheumatoid arthritis, hypothyroidism, idiopathic thrombocytopenic purpura, hypertension, and migraines. She reported hitting her left shin on a rock in August 2020 while participating in a recreational water activity in a south-central Pennsylvania creek (Figure 1). The patient reported redness and discomfort developing in the area the day after injury and severe pain of the lower leg around superficial abrasions on the left shin that developed 3 days later. The patient subjectively reported low-grade fever, without objective temperatures >100°F (37.8°C). Four days after the initial exposure, the patient was evaluated by her primary care physician, who was concerned that compartment syndrome might have developed and advised her to go to the ED for further evaluation.

During ED examination, the patient reported severe left lower leg pain. Temperature was 37.7°C, heart rate 121 beats/min, respiratory rate 18 breaths/min, blood pressure 150/108 mm Hg, and oxygen saturation 100% on room air. Physical examination revealed erythema and tenderness of the left lower extremity, and a 1 cm abrasion was located on the anterior aspect of the leg distal to the left knee. Edema of the left lower leg was also noted. The rest of the physical examination was unremarkable. Laboratory results at admission revealed an elevated leukocyte count of 24,200 cells/μL (reference range 3,900–9,500 cells/μL) (Table).

The patient was hospitalized for sepsis attributable to cellulitis and administered a single dose of IV ceftriaxone (1 g) in the ED and started on IV cefazolin antimicrobial therapy (2g every 8 h) after admission as treatment for potential group B Streptococcus infection. Radiographs of the patient’s left lower fibula and tibia showed subcutaneous phleboliths (calcification of blood clots) likely unrelated to the injury and no radiopaque foreign body or acute osseous abnormalities. The patient clinically improved during hospitalization with decreased left leg pain, improved leukocyte count, and improved physical therapy evaluations. She was discharged after 2 days and prescribed an 8-day course of oral cephalexin (500 mg 4×/d).

Within 2 days of discharge, the blood cultures performed at admission were positive for unspecified gram-negative rod-shaped bacteria, and the patient was advised to return to the ED for evaluation. At admission, the patient reported improved symptoms with decreased redness and swelling at the site and an increased ability to ambulate on her left leg. She denied any fevers, chills, nausea, or vomiting. On physical examination, a slight warmth around the healing abrasions was noted, the erythema on their left leg had improved, and no edema was apparent. Repeat blood culture, lactic acid, basal metabolic panel, and complete blood count were performed. Other than a slightly elevated leukocyte count at 9,600 cells/μL (reference range 3,900–9,500 cells/μL), other values were unremarkable. The patient was hospitalized and treated with IV cefepime (2 g every 8 h); while hospitalized, the patient continued to clinically improve. She was discharged on day 3 with a 10-day course of oral levofloxacin (750 mg/d). The blood cultures from blood obtained during the second hospitalization were negative. Two months later, we identified the isolate from a blood sample obtained during the patient’s initial hospitalization as H. jordaniae (GenBank accession no. PX060171). A nearly full-length (1,408-bp) 16S rRNA gene sequence from the isolate shared 99.9% sequence similarity with DSM 28903T.

Patient 3

Patient 3 was a 36-year-old man with history of diabetes mellitus, obesity, and boils. He reported hitting his leg in July 2020 while participating in a recreational water activity in a south-central Pennsylvania river (Figure 1). The patient sought care at the ED the day after the injury with pain, swelling, and erythema of the anterior aspect of the right leg.

At initial examination, the patient reported severe right lower leg pain. Temperature was 36.4°C, heart rate 106 beats/min, respiratory rate 16 breaths/min, blood pressure 164/76 mm Hg, and oxygen saturation 97% on room air. Physical examination showed a 10.5 cm area of poorly defined erythema, warmth, and tenderness of the anterior aspect of the right leg. No drainage, breakage of the skin, or crepitus of the limb’s joints was noted. Laboratory results at admission demonstrated an elevated leukocyte count of 22,430 cells/μL (reference range 4,000–10,800 cells/μL) (Table).

The patient was administered a single dose of IV clindamycin (800 mg) for presumed bacterial infection. Radiographs of the right tibia and fibula showed chronic calcification of the diaphysis of the tibia, unchanged from a prior radiograph, and no signs of acute osseous abnormalities, radiopaque foreign body, or soft tissue gas. On the basis of the patient’s age, comorbidities, laboratory results, imaging, and examination findings, the decision to treat as an outpatient with a primary care physician follow-up was made. The patient was discharged with a 10-day course of oral clindamycin (300 mg every 8 h).

Within 4 days of discharge, the blood culture performed at admission was positive for gram-negative rod-shaped bacteria. The patient was advised to return to the ED for evaluation and additional blood cultures. At admission, the patient reported feeling better with right leg improvement. He had slight tenderness and itchiness of the right leg, with pain only when walking and during palpation. Physical examination showed an increased warmth of the right anterior lower leg compared with the left leg and a central scabbed-over pinpoint wound on the right leg with mild erythema. Bloodwork revealed resolved leukocytosis; other blood values, radiographs, and urinalysis results were unremarkable. The patient was hospitalized and received IV cefepime (dose unknown) for bacteremia and cellulitis of the right lower extremity. While hospitalized, the patient clinically improved, and the redness of the right leg completely resolved. Three days after collection, preliminary results for the repeat blood cultures were negative. After consultation with an infectious disease doctor, the patient was discharged on a course of oral ciprofloxacin (500 mg every 12 h) and oral amoxicillin/clavulanate (500 mg/125 mg every 8 h) (length of course unknown). The blood cultures obtained during hospitalization were negative after 5 days of incubation. Gram-negative rod-shaped bacteria were recovered by the hospital laboratory from blood cultures collected at the patient’s initial ED visit. We identified the isolate as H. jordaniae 1.5 months after the patient’s discharge (GenBank accession no. PX060170). A nearly full-length (1,408-bp) 16S rRNA gene sequence from the isolate shared 99.9% sequence similarity with DSM 28903T.

Patient 4

Patient 4 was a 48-year-old man with history of fatty liver who sought care at the ED in July 2020 with a history of a fever up to 101°F (38.3°C), nausea, and left lower leg erythema with swelling of 1-day duration (Figure 1). The patient reported being at a cabin the prior 2 days and noticed bug bites and scratches develop on his legs while walking in shorts. Additional information about any water-related exposures for this patient are unknown as the patient was unable to be reached for an interview.

At initial examination, the patient had a fever of 38.6°C; heart rate was 111 beats/min, blood pressure 107/67 mm Hg, and an oxygen saturation of 97% on room air. Physical examination showed bug bites and erythema on the anterior distal shin of the left leg that did not extend to the knee or thigh. Mild edema was also noted in areas of erythema. Tachycardia was the only other major physical examination finding. Laboratory results at admission revealed an elevated leukocyte count of 21,000 cells/μL (reference range 4,000–11,000 cells/μL) (Table).

The patient was observed in hospital for cellulitis and administered a single dose of IV cefazolin (2g). Ultrasound of the left lower extremity revealed no signs of deep vein thrombosis. The patient’s fever and tachycardia improved with IV fluids, ketorolac tromethamine, and acetaminophen. The patient was discharged ≈4 hours after admission with a 7-day course of oral sulfamethoxazole/trimethoprim (800 mg/160 mg every 12 h). Two months later, we identified an isolate from a blood sample obtained during the patient’s ED treatment as H. jordaniae (GenBank accession no. PX060172). A nearly full-length (1,408-bp) 16S rRNA gene sequence from the isolate shared 99.9% sequence similarity with DSM 28903T.

Source