Wound Drainage After Trauma in a Fourteen-Year-Old Girl

A 14-year-old girl was struck by a car while waiting at a bus stop. On arrival in the emergency department, she had a temperature of 36.7°C, pulse 82 beats/min, respiratory rate 18 breaths/ min, and blood pressure 90/53 mm Hg. Physical examination was normal apart from tenderness of the left elbow and both knees, with a surface abrasion over the right knee and a deep laceration over the left knee. Radiographs demonstrated a right bicondylar tibial plateau fracture. She had been treated 1 year previously for ethanol poisoning but had otherwise been healthy. She was living in a group home because of a conduct disorder. External fixation of the right tibial plateau fracture and debridement of the soft tissue injury to the left knee were performed 3 days after the trauma. Open reduction and internal fixation of the fracture was then performed 3 days later, and examination under anesthesia demonstrated tears of the left posterior cruciate and medical collateral ligaments. Two days after this second surgery, she had fever to 39.4°C which persisted for the next week, and the right knee became progressively more swollen with the incision becoming red and purulent. Gram stain of the discharge showed 3+ polymorphonuclear leukocytes, 4+ red blood cells and no bacteria, and cultures were sterile. Treatment with cephalexin was started 9 days after the initial trauma and changed to cefazolin the next day. Despite this treatment, the purulent drainage from the right knee incision persisted. A complete blood count revealed a white blood cell count of 13,800/mm3 with 81% neutrophils, 11% lymphocytes, 7% monocytes and 1% eosinophils. Hemoglobin was 7.7 g/dL and platelets were 648,000/mm3. Erythrocyte sedimentation rate was 72 mm/h. Urinalysis showed 2+ leukocytes and >50 white blood cells per high power field. Urine culture was not done, and blood culture was sterile. Arthrotomy and synovectomy were performed 15 days after the initial trauma. Copious purulent material was noted in the joint and bathing the tibial plate. A Gram-stained smear of this material showed 2+ red blood cells, 1+ polymorphonuclear cells, 1+ mononuclear cells and no bacteria. A white blood cell scan pre-formed the day after the synovectomy showed increased uptake in the soft tissues of the left knee, and at the screw placement sites on the right tibia. Repeat debridement of the right knee was performed 2 days after the synovectomy, which showed minimal purulent material. Six days after the synovectomy, an organism was isolated from the synovial fluid. Denouement Mycoplasma hominis grew on brain-heart infusion media (enriched with 5% sheep blood, vitamin K and hemin) from the synovial fluid. The patient completed a further 4-week course of cefazolin followed by a 1-week course of cephalexin, given concurrently with a 5-week course of oral doxycycline for M. hominis septic arthritis. The incision became tender 6 weeks after completion of that antibiotic therapy, but this resolved with a 1-week course of oral clindamycin. The patient was well when last assessed 2 months after completing the clindamycin. M. hominis is part of normal genital flora and is most commonly acquired through sexual transmission. 1 Colonization is usually asymptomatic but has been linked to pyelonephritis, pelvic inflammatory disease, and postpartum or postabortal fever. 1 Risk factors for M. hominis septic arthritis include immunosuppression (especially hypogammaglobulinemia), trauma, childbirth and urogenital manipulation. 2 Isolation of M. hominis from the synovial fluid of 1 child with juvenile rheumatoid arthritis and 2 children with a diagnosis of septic arthritis has been described, 3 so it appears that septic arthritis can also occur in the absence of risk factors. The pathogenesis of the septic arthritis is presumably seeding of the joint during bacteremia, rather than local spread from the genital tract. Bacteremia with M. hominis has been documented after trauma and childbirth and in patients with severe chronic disease 4 and could be a common phenomenon that is rarely detected, because the mycoplasmas might not produce turbidity in the blood culture vial. 2 The expected time between the bacteremia and the onset of septic arthritis has not been delineated. Our patient had signs of infection 8 days after the trauma. Entry of the organism in the blood stream could have occurred at the time of the trauma, or related to one of her later surgical procedures. She had no urinary catheter, which could also predispose patients to bacteremia with M. hominis. 2 No investigations were done in our patient to exclude osteomyelitis, but the patient received a duration of therapy that would likely be adequate for osteomyelitis. However, osteomyelitis caused by M. hominis seems to always involve the sternum. 5 There is increasing resistance of M. hominis to doxycycline; therefore clindamycin is a better empiric choice for M. hominis septic arthritis. 2 The minimum duration of therapy is not clear. Eleven of 14 reported cases reported were cured by 2–17 weeks of antibiotic therapy, and the 3 recurrences were in immunocompromised hosts treated for 3–7 months. 2 Drainage of the joint and decreasing immunosuppression have been recommended as part of therapy. 2 In 4 previous reports of infection in the presence of prosthetic material, 3 were cured with a 4- to 10-week course of antibiotics, 2,6,7 but another case recurred after 8 months of treatment with doxycycline and was eventually suppressed with long term use of ciprofloxacin. 8 Our patient’s infection was cured without removal of the tibial plate. Bacteremia with M. hominis can be self-limited in the postpartum period, 9 but it seems less likely that septic arthritis from M. hominis is ever self-limited or cured by debridement alone. Our patient became afebrile before the use of doxycycline. The final debridement done 2 days after the synovectomy, and 4 days before doxycycline was added, showed marked improvement. It is not clear whether the infection would have eventually progressed in the absence of appropriate antimicrobials. Making the diagnosis of M. hominis septic arthritis is difficult. There are no distinguishing features for Mycoplasma infection on analysis of synovial fluid. 2 Mycoplasmas do not Gram stain and grow as pinpoint, flat translucent colonies on routine medium 2; therefore they can easily be missed on culture. However, synovial fluid cultures are sterile in ~40% of cases of septic arthritis, 10 so sterile cultures alone are not a useful clue. The diagnosis of M. hominis septic arthritis should be considered and synovial fluid cultured for mycoplasma if routine cultures are sterile and the patient is not responding to appropriate antibiotics, especially if the patient is immunosuppressed or if the septic arthritis follows trauma or childbirth.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.