 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Toxic Shock Syndrome Complicating Orthopaedic Manipulation of Bone. A Report of Two Cases*  , CHICAGO, ILLINOISInvestigation performed at Wyler Children's Hospital, The University of Chicago, and Children's Memorial Hospital, Northwestern University, Chicago
In 1978, Todd and Fishaut described toxic shock syndrome, a multisystem disorder usually associated with toxic shock syndrome toxin-1, a toxin elaborated by Staphylococcus aureus. Most cases occur during menstruation and are associated with vaginal colonization by strains of this bacterium that elaborate the toxin. Other risk factors include the use of tampons, a young age, and a low serum concentration of anti-toxic shock syndrome toxin-1 antibody8. Toxic shock syndrome can also occur in patients who are not menstruating; the risk factors are less certain but include nasal operations and the presence of a non-operative or operative wound. Toxic shock syndrome that is not related to menstruation is less often associated with the isolation of Staphylococcus aureus that produces toxic shock syndrome toxin-1; only 40 to 60 per cent of isolates from such patients produce the toxin, compared with 90 to 100 per cent of isolates from patients who are menstruating4,6,14. Other toxins, most notably staphylococcal enterotoxin-B, have been implicated in toxic shock syndrome in non-menstruating patients. Orthopaedic procedures have not been considered a risk factor for toxic shock syndrome, and there have been few reports of toxic shock syndrome associated with bone manipulation and implants. One report described two patients who survived after insertion of an external fixator18 and one, a patient who died after removal of a metallic internal-fixation device from the healed site of a right femoral varus derotational osteotomy combined with a right arthrodesis of the wrist with iliac-crest bone graft and fixation with two Kirschner wires13. Staphylococcus aureus grew on culture of specimens obtained during incision and drainage of the iliac crest and the wrist13. Additionally, Irvine et al. reported a death due to toxic shock syndrome after removal of three Steinmann pins and osteoplasty of a healed slipped capital femoral epiphysis. In both fatal cases, the diagnosis of toxic shock syndrome had not been considered until late in the postoperative course because the operative wound appeared uninfected. We recently managed two children who had toxic shock syndrome after open fixation for a slipped capital femoral epiphysis. These cases demonstrate the importance of considering this diagnosis early in the postoperative course of orthopaedic patients in whom a septic, shock-like syndrome develops. They also serve to remind clinicians that toxin-producing organisms may be present despite the absence of clinically apparent infection.
CASE 1. A premenarchal twelve-year-old girl had open fixation of a grade-I slipped capital femoral epiphysis with a single cannulated screw through a formal incision. No antibiotics were given perioperatively. Seventy-two hours later, fever (40.5 degrees Celsius), hypotension, and diarrhea ensued, and the patient was transferred to a tertiary-care hospital. Physical examination on arrival revealed an acutely ill, disoriented child. The temperature was 38.9 degrees Celsius, the pulse rate was 148 beats per minute, and the blood pressure was 85/60 millimeters of mercury (11.33/8.00 kilopascals). Progressive macular erythroderma developed over the lower portion of the abdomen, and conjunctival erythema was noted. The lungs were clear to auscultation; there was no cardiac murmur. Hepatosplenomegaly was absent. Examination of the extremities revealed normal findings; the operative site was clean, without erythema or tenderness. Laboratory studies revealed normal levels of blood urea nitrogen, serum creatinine, aspartate aminotransferase, alanine transferase, and total bilirubin, but the platelet count was 90,000 per cubic millimeter (90 x 109 per liter). Because of hypotension, rash, and diarrhea, a diagnosis of toxic shock syndrome was suspected, which prompted incision and drainage of the hip wound. Staphylococcus aureus grew on culture of a specimen from the wound and on culture of a specimen from the nasopharynx. The patient responded to fluid resuscitation and intravenous administration of cephapirin but remained febrile, with a temperature of about 39 degrees Celsius. Because of the persistent fever, two open débridements were performed during the next five days. No abscess or purulent material was found. The fever continued, and the screw was removed ten days after the original operative procedure. Forty-eight hours later, the patient was afebrile and was discharged. One week later, desquamation of the rash occurred. The patient was readmitted four weeks later for elective, percutaneous fixation through a separate incision; clindamycin was administered preoperatively and postoperatively. The contralateral slipped capital femoral epiphysis was fixed percutaneously six months later; vancomycin and rifampin were administered prophylactically during the perioperative period. Both procedures were well tolerated. CASE 2. An eleven-year-old girl had an acute slipped left capital femoral epiphysis in April 1991, and fixation with four Knowles pins was performed at a community hospital. Cefazolin was administered perioperatively. Three weeks postoperatively, nausea, vomiting, and diarrhea developed; the patient was seen forty-eight hours later in an emergency room for fever, shock, and transient loss of consciousness. The temperature was 40 degrees Celsius. The pulse rate was 150 to 160 beats per minute, and the blood pressure fell acutely to 70/50 millimeters of mercury (9.33/6.67 kilopascals). The patient was transferred to a pediatric intensive-care unit where she received fluid resuscitation, dopamine, and dobutamine. Assisted ventilation was necessitated by progressive respiratory failure. The physical examination revealed a macular, erythematous rash on the left palm, right hip, and middle portion of the abdomen. Rales were heard bilaterally on auscultation of the lung fields. The cardiac examination revealed a gallop rhythm and systolic murmur that was grade II of VI. The abdomen was diffusely tender, with hypoactive bowel sounds; hepatosplenomegaly was absent. The hip appeared normal, and the site of the operative wound lacked erythema or tenderness. A radiograph of the chest showed a diffuse, reticular-nodular pattern, consistent with adult respiratory-distress syndrome. Ultrasonographic examination revealed a normal abdomen. The patient was managed empirically with vancomycin, gentamicin, and metronidazole for presumptive septic shock. A diagnosis of toxic shock syndrome was considered the next morning. The patient remained febrile for six days. Two punch biopsies of the skin revealed mild perivascular lymphohistiocytic infiltrate consistent with toxic erythema. The rash became progressively erythematous and subsequently desquamated. Laboratory studies showed mildly increased serum concentrations of aspartate aminotransferase, alanine tranferase, blood urea nitrogen, and creatinine. Proteinuria (3+) was documented. Computed tomography of the abdomen revealed moderate ascites, which, after paracentesis, was found to be a transudate. A gallium-67 scan showed increased uptake in the left hip, consistent with osteomyelitis or previous operative manipulation. The patient received vancomycin and gentamicin intravenously for fourteen days but, because the hip could have been the source of the toxic shock syndrome, a decision was made to administer a prolonged course of antibiotics orally for possible Staphylococcus aureus osteomyelitis. The operative pins were not removed at this time because of the uncertainty of the diagnosis and the mechanical necessity of the pins. The patient was discharged while receiving dicloxacillin, until the pins could be removed with minimum risk to the joint. Interestingly, a Staphylococcus aureus isolate, obtained from the oropharynx, was subsequently found to elaborate enterotoxin-B and not toxic shock syndrome toxin-1 (Associated Regional and University Pathologists, Salt Lake City, Utah). No anti-enterotoxin-B antibody was detected in a serum sample obtained on the sixth day of the illness; however, anti-toxic shock syndrome toxin-1 antibody was present (Associated Regional and University Pathologists). About two months later, radiographs revealed collapse of the femoral head and evidence that the four Knowles pins were beginning to encroach into the joint. The patient was hospitalized for removal of the pins. Vancomycin was administered prophylactically before the operation. A culture of a specimen obtained from aspiration of the hip preoperatively revealed no growth. The patient received two doses of cefazolin postoperatively and was discharged. Twenty-four hours later, the patient returned with fever, nausea, vomiting, and progressive hypotension. The temperature was 41 degrees Celsius, the pulse rate was 140 to 150 beats per minute, the respiratory rate was thirty breaths per minute, and the blood pressure was 114/37 millimeters of mercury (15.20/4.93 kilopascals). There were no rashes. Examination of the left hip revealed a fluid collection over the greater trochanter. The incision in the left hip was reopened operatively, and about thirty milliliters of blood-tinged fluid was obtained. A gram stain revealed no microorganisms, and the culture of the fluid was sterile. The wound was packed open. Radiographs of the hip with use of indium scintigraphy showed increased uptake at the femoral head and neck, consistent with osteomyelitis or operative manipulation. The patient was managed with vancomycin intravenously for eight days and was discharged while receiving dicloxacillin orally. She continued to take the dicloxacillin for presumptive chronic osteomyelitis for three months, without any additional complications.
The association between toxic shock syndrome and internal fixation devices and the need to consider the prompt removal of such devices when toxic shock syndrome is suspected have not been adequately emphasized in the orthopaedic literature. The cases of our two patients suggest that, in contrast to osteomyelitis, which usually resolves after antibacterial therapy and drainage, toxic shock syndrome may also necessitate removal of internal devices. This seems to have been particularly underscored by the case of our first patient (Case 1), in whom the fever did not subside, despite appropriate antibiotic therapy, drainage, and other supportive measures, until the pins had been removed. Our second patient (Case 2) responded well to medical management without removal of the pins. However, when the pins were electively removed two months later, the toxic shock syndrome recurred. We presume that either manipulation of the pins or intubation for the elective operation triggered the recurrence. The clinical features of toxic shock syndrome have been described previously2,17. Several that are associated with operative procedures deserve emphasis and may be helpful in making the diagnosis. First, the operative site may not appear infected, as was true in both of our patients. Second, toxic shock syndrome may have a short incubation period. Seven of thirteen non-menstruating patients1 had onset of the symptoms within forty-eight hours after an operative procedure, an interval similar to that in both of our patients. Third, the disease may progress rapidly. The diagnosis of toxic shock syndrome depends on clinical judgment as there is no definitive laboratory test. In patients who are not menstruating, the isolation of Staphylococcus aureus from a normally sterile site in the body (for example, from the hip in our Case 1) supports the diagnosis. The case of our second patient (Case 2) demonstrates the complexities involved in the diagnosis of toxic shock syndrome. Staphylococcus aureus was grown on culture of a specimen from the throat during the first episode of toxic shock syndrome, although specimens were not obtained from the site of the hip operation. When the toxic shock syndrome recurred, material obtained from the hip was sterile on culture, perhaps as a result of the long course of antibiotics. Isolation of toxin-producing Staphylococcus aureus from a mucous membrane of a patient who has clinical features of toxic shock syndrome suggests a diagnosis of toxic shock syndrome, but such isolates are also found in asymptomatic individuals. It has been estimated9-12 that as much as 35 per cent of the population has Staphylococcus aureus in the nasopharynx and 10 per cent of the isolates elaborate toxic shock syndrome toxin-1. Antibodies to toxic shock syndrome toxin-1 are present in the serum of more than 90 per cent of individuals who carry toxic shock syndrome toxin-1-positive strains of Staphylococcus aureus9. It is unclear whether the Staphylococcus aureus isolated from the oropharynx of our second patient (Case 2), which did not elaborate toxic shock syndrome toxin-1, was responsible for the disease or whether a different isolate was present at the operative site or elsewhere. The recurrence of toxic shock syndrome following operative manipulation of the pins suggests that the hip harbored the responsible organism. However, because the child was intubated for the elective removal of the pins, the possibility that the isolate from the throat was the source cannot be excluded. Staphylococcus aureus isolates associated with toxic shock syndrome may produce toxins other than toxic shock syndrome toxin-1. Examples of such toxins14 include enterotoxin-B, elaborated by the isolate from the throat of our second patient (Case 2), or enterotoxin-C. Interestingly, isolates of Streptococcus pyogenes have also been associated with a toxic shock syndrome-like illness15 and often elaborate streptococcal pyrogenic exotoxin-A. Moreover, there is substantial homology between streptococcal pyrogenic exotoxin-A and Staphylococcus aureus enterotoxins B and C, although toxic shock syndrome toxin-1 does not share this homology11. It is believed that all of these toxins mediate a toxic shock syndrome-like illness because they behave as superantigens and stimulate T-cells expressing specific V-ß chains5,11,19. Such activation of T-cells with resultant release of various lymphokines is presumed to result in toxic shock syndrome. The assaying of antitoxin antibody in serum is of limited use in the diagnosis of toxic shock syndrome. While an increase in serum antibody to toxic shock syndrome toxin-1 in association with clinical manifestations of the disease provides strong retrospective support for the diagnosis, at least one-third of toxic shock syndromes in non-menstruating patients are due to strains of Staphylococcus aureus that do not produce toxic shock syndrome toxin-1. Also, the lack of protective antibody is an important risk factor for toxic shock syndrome. The relatively rare occurrence of toxic shock syndrome in the population carrying organisms producing toxic shock syndrome toxin-1 in the mucous membranes appears to be attributable, in large part, to protective levels of antibody to toxic shock syndrome toxin-1. Current recommendations for the treatment of toxic shock syndrome include intravenous administration of an antistaphylococcal antibacterial agent and drainage and irrigation of infected sites2. The cases of our two patients underscore the need to consider removal of potentially colonized foreign bodies. A possible role for intravenous administration of gamma globulin has also been advocated by some investigators3,10, although prospective, controlled clinical trials have not been performed, to our knowledge. Most preparations of immunoglobulin for intravenous administration have substantial concentrations of anti-toxic shock syndrome toxin-1 antibody. In animal models of toxic shock syndrome, the intravenous administration of human gamma globulin (one gram per kilogram of body weight), given simultaneously with toxin or toxin-producing organisms, has prevented toxic shock syndrome3,10.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Introduction
Case Reports| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |