Clostridium species are capable of producing several types of infectious processes, many of which have proven to be life-threatening. Septic arthritis caused by Clostridium, however, is not a very frequent finding. Currently, only 37 cases of infectious arthritis due to Clostridium species have been reported. We report a case of septic arthritis in which Clostridium perfringens, Clostridium sordellii, and Clostridium tertium were each isolated from the synovial aspirate. In addition, the 37 previously reported cases are summarized to compare the similarities and differences of the clinical course, treatment, and outcome, in order to help establish guidelines for the proper management of this infectious process.
Septic arthritis caused by anaerobic organisms is a rare clinical entity. Anaerobic septic arthritis accounts for only 1% of all reported cases of bacterial arthritis in both children and adults . Rarely has a Clostridium species been reported as the pathogenic agent of infectious arthritis. We reviewed the English-language literature from 1966 to 1999 by means of MEDLINE and found only 37 cases of septic arthritis caused by a Clostridium species [2-27]. In each of these cases, only 1 species of Clostridium was the causative agent. We review these 37 cases and report a case of septic arthritis in which Clostridium perfringens, Clostridium sordellii, and Clostridium tertium were each isolated from the synovial aspirate.
A 37-year-old healthy man was seen in an emergency department ∼24 h after sustaining a puncture wound to the medial aspect of the left foot. A metal fragment penetrated his leather boot as he was pushing trash with his foot. Throughout the night, the pain progressively worsened, and upon awakening the next morning, he noticed that the foot was red and swollen, at which time he sought medical treatment. The patient's medical history was insignificant. He was taking no medications, and he had no known drug allergies.
At the time of physical examination, the patient's oral temperature was 37°C, his pulse rate was 91 and regular, his respiration rate was 16, and his blood pressure was 156/70 mm Hg. Examination of the left foot revealed moderate edema, warmth, and erythema about the medial aspect of the forefoot. A small puncture wound was evident at the medial aspect of the first metatarsophalangeal joint from which no drainage could be expressed (figure 1). There was severe tenderness of the first metatarsophalangeal joint to palpation and with range of motion. Neurovascular status was intact to the left lower extremity.
Radiographs showed no fracture, dislocation, or presence of a foreign body. Increased soft-tissue density was seen about the first metatarsophalangeal joint with a possible joint effusion. There was no evidence of bony erosion, joint space narrowing, or gas in the joint or surrounding subcutaneous soft tissues.
Laboratory studies revealed a total WBC count of 9200/mm3, with an erythrocyte sedimentation rate of 6 mm/h. Aspiration of the joint in the emergency department and again the following day yielded ∼2 mL of sanguinopurulent fluid that was extremely cloudy and did not clot. Cultures of both fluid samples subsequently yielded C. tertium susceptible to penicillin.
The patient was treated with penicillin G (4 × 106 U every 4 h) and iv clindamycin (600 mg every 6 h). One day later, the patient remained symptomatic with pain and limited joint range of motion, warranting open arthrotomy in the operating department. Intraoperatively, no purulent drainage could be found in the subcutaneous tissues surrounding the joint. However, when the joint capsule was incised, ∼4 mL of purulent drainage flowed freely from the joint (figure 2). The synovia was inflamed, hypertrophic, and markedly edematous. The cartilaginous surfaces of both the first metatarsal head and the base of the proximal phalanx appeared grossly normal without erosions or lytic lesions (figure 3). The joint was irrigated with normal sterile saline by means of pulse lavage, and the wound was packed open. Cultures of specimens obtained intraoperatively yielded C. perfringens, C. sordellii, and C. tertium.
After the operations, the wound was irrigated and the sterile dressing changed daily. The patient remained afebrile and continued to show signs of improvement. On the seventh day after the operation, the patient underwent delayed primary closure and was discharged; treatment at the time of discharge included a 2-week course of oral clindamycin and no bearing of weight on the left lower extremity. Erythema and edema continued to decrease, and the joint regained function equal to that of the corresponding joint in the opposite foot.
Clostridium species have been implicated as the causative agent of a diverse variety of soft-tissue and alimentary infections, including myonecrosis, fulminating cellulitis, and food poisoning. Septic arthritis due to Clostridium species is an unusual clinical occurrence that is rarely documented. The 37 previously reported cases of septic arthritis caused by a Clostridium species are summarized in table 1. Although a hematogenous route has been reported, most cases were associated with a traumatic, penetrating injury. C. perfringens was the most common infecting species. In most instances, Clostridium was the sole pathogenic factor; however, 5 polymicrobial infections were reported. The knee was the most frequent joint involved, and in no case reviewed was a joint in the foot implicated.
The diagnosis was established through isolation of the bacteria from a synovial fluid aspirate in all of the reported cases. Typically, the patient presented afebrile or with a low-grade fever. Gas in the subcutaneous tissues, commonly seen in anaerobic infections (particularly involving C. perfringens), was evident only in 5 of the 37 cases. Examination of the involved joint usually revealed erythema, calor, and a joint effusion. Joint range of motion was most often painful and limited.
In all cases in which the joint was inspected surgically, the synovia was found to be edematous, inflamed, hypertrophic, or necrotic. There was cartilaginous destruction in only 2 of the reported cases; both of these patients had preexisting arthritic conditions. Therefore, it is difficult to conclude that the infectious process was the cause of the joint destruction. Typically, clostridial infections result in necrosis and gross tissue destruction with or without subcutaneous gas. Because of the toxic enzymes that Clostridium species are known to release, cartilaginous destruction might be the expected result of a clostridial joint infection. It is interesting that this result was not found to be true in either the previously reported cases or our case.
Management of general septic arthritis has been thoroughly described in the literature; however, treatment of clostridial septic arthritis is not well outlined. Some investigators advocate serial aspirations of the joint space but emphasize the necessity of surgical synovectomy in any case that is not responsive after several days. In the cases reviewed, most of the patients underwent open arthrotomy. In all cases, penicillin G was the antimicrobial agent used once the pathogenic organism was determined to be Clostridium.
When addressed early, the treatment of clostridial septic arthritis has produced satisfactory results. With the exception of immunocompromised persons, successful results were obtained in all cases. A review of the literature has revealed that the treatment of choice seems to be iv penicillin and open arthrotomy performed somewhat soon after the time of onset. With prompt diagnosis and proper management, a good functional result can be expected.
The case reported here is similar in many aspects to the previously reported cases. The infectious process was traumatically induced like in most of the other reported cases, with the exception of hematogenous inoculation, which is most often seen in immunocompromised and pediatric patients. Except for the synovial inflammation, no joint destruction was evident, which was also true in all but 2 of the reported cases. In conclusion, infectious arthritis caused by Clostridium is a rare clinical occurrence. To our knowledge, the isolation of 3 different Clostridium species from synovial fluid has never been documented. However, review of the literature has proven that a good functional result can be expected given an early diagnosis and proper management.