 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Comparison of Intravenous and Oral Antibiotic Therapy in the Treatment of Fractures Caused by Low-Velocity Gunshots. A Prospective, Randomized Study of Infection Rates*  , LOS ANGELES, CALIFORNIAInvestigation performed at the Orthopaedic Trauma Service, Los Angeles County-University of Southern California Medical Center, Los Angeles
One hundred and ninety consecutive patients (222 fractures) who had an extra-articular fracture of a long bone as a result of a low-velocity gunshot were randomized into two groups on the basis of the method of administration of antibiotics. Group 1 consisted of 101 patients (120 fractures) who were managed with intravenous administration of cephapirin sodium and gentamicin for three days. Group 2 comprised eighty-nine patients (102 fractures) who were managed with oral administration of ciprofloxacin for three days. The two groups were comparable in terms of the age of the patient, the locations of the fractures, and the time from the injury to the commencement of antibiotic therapy. Injuries that needed operative débridement or fixation were excluded. All patients were followed until the fracture had healed. Two infections developed in two of the ninety-nine patients (118 fractures) who completed the study in Group 1, and two infections developed in two of the eighty-seven patients (100 fractures) who completed the study in Group 2. With the numbers available, there was no significant difference in the rates of infection (2 per cent for both) between the two groups. All four fractures that were complicated by infection were located in the distal half of the tibia. We concluded that oral and intravenous administration of antibiotics were equally effective for prophylaxis against infection after an extra-articular fracture from a low-velocity gunshot.
The number of fractures caused by firearms increases yearly25. There are approximately 7.5 non-fatal injuries for each fatal gunshot injury (range, 3:1 to 105:1)25. In 1985, gunshots caused nearly 268,000 injuries, including 31,556 that were fatal, 65,127 that led to hospitalization, and 171,000 that needed some outpatient medical care or resulted in at least one day lost from usual activity5,6,25. The direct and indirect costs in 1985 were estimated to be 14.4 billion dollars5,6,25. The cost of providing medical care for these patients has placed a financial strain on medical care providers and insurers18,24,27,31. Wartime experiences have provided the scientific basis for the proper treatment of gunshot wounds7,8,19,29. There is little disagreement regarding the treatment of high-velocity gunshot wounds (operative exploration, extensive primary débridement, treatment of the open wound, and intravenous administration of antibiotics)1-3,10,14,15,21,32,35. However, major controversy remains regarding the role of antibiotics in the treatment of fractures caused by low-velocity gunshots. In the literature, recommendations have ranged from no antibiotic therapy to the intravenous administration of antibiotics for three days9,11,17,23,28. The purpose of the present study was to determine the rates of infection associated with oral and intravenous administration of antibiotics in the treatment of extra-articular fractures of long bones due to low-velocity gunshots.
All patients who were seen in the emergency room at the Los Angeles County-University of Southern California Medical Center between October 1, 1992, and December 31, 1993, for a fracture of a long bone from a low-velocity gunshot were admitted to the Orthopaedic Trauma Service and were evaluated for inclusion in the study. The criteria for exclusion were an intermediate or high-velocity-gunshot wound, as determined on the basis of the history and the characteristics of the wound; skeletal immaturity; a wound that was one centimeter in diameter or more; a vascular injury; pregnancy; a fracture involving a foot or a hand; and a gunshot wound that traversed the abdomen. Injuries that needed débridement, fixation of the fracture, or another operative procedure in which the patient received antibiotics perioperatively were also excluded. The three senior ones of us (T. P. K., M. J. P., and J. L.) evaluated the patients to determine if they met the criteria for inclusion in the study. A signed informed-consent form was obtained from all of the patients, and the patients were randomized by medical-record number into two treatment groups. All of the patients were hospitalized during the initial three-day period to ensure compliance with the treatment protocol. The patients in Group 1 were managed with intravenous administration of cephapirin sodium (two grams every four hours) and gentamicin (eighty milligrams every eight hours) for three days. The patients in Group 2 were managed with oral administration of ciprofloxacin (750 milligrams twice daily) for three days. One hundred and ninety-eight male patients and four female patients were admitted into the study. Eight patients (ten fractures) in Group 1 and four patients (four fractures) in Group 2 chose not to participate in the study and therefore received our current treatment protocol, which was the same as that used for Group 1. These twelve patients were not included in the study but were followed for the sake of completeness. Of the remaining 190 patients (222 fractures), 101 (120 fractures) were in Group 1 and eighty-nine (102 fractures) were in Group 2. The average age at the time of injury was twenty-five years (range, fifteen to fifty-seven years). The average age was twenty-five years (range, fifteen to fifty-seven years) in Group 1 and twenty-six years (range, sixteen to fifty-two years) in Group 2. The two groups were also comparable in terms of the anatomical location of the fractures (Table I) and the time from the injury to the commencement of antibiotic therapy.
One patient (one fracture) in Group 1 was lost to follow-up. No patients in Group 2 were lost to follow-up. One patient (one fracture) in Group 1 and two patients (two fractures) in Group 2 were removed from the study because they had received prophylactic perioperative antibiotics. One of these three patients was incarcerated after the injury, and the physician at the prison thought that external fixation without open reduction of the tibial fracture was indicated. Another patient later sustained a fracture of the ankle that was treated with operative intervention. The third patient had an additional gunshot injury several weeks after the initial injury and needed an emergency celiotomy. All three fractures healed without infection. In the emergency room, treatment included superficial débridement of the entrance wounds and, if present, the exit wounds as necessary. (With use of local anesthesia, the wound was cleaned with Betadine [povidone-iodine] surgical scrub and foreign material was removed.) The fracture site was immobilized in whatever manner was appropriate. All extremities were evaluated for neurovascular injuries by physical examination and, if necessary, by Doppler ultrasonography. No patient in the present study needed angiography. Compartment pressures were measured, if necessary. According to the criteria for exclusion, no patient who had a fasciotomy was included in the study. Wounds were left open to heal by granulation. All of the patients received prophylaxis against tetanus. Treatment was considered to have failed if signs or symptoms of infection developed, and the infection was treated appropriately. We adhered to a strict definition of infection, with one exception. It has been our experience that, with most gunshot injuries, a red ring rapidly develops around the wound secondary to the heat of the projectile. This so-called bullet cellulitis was not considered a sign of infection. After discharge from the hospital, all of the patients in the study were followed at regular intervals at a special clinic, where radiographs were made and the patients received wound care and fracture management. The patients were followed until there was both clinical and radiographic evidence of union. The healing time for the fracture was defined as the interval between the injury and the time when the patient had clinical and radiographic evidence of complete healing.
The rate of infection was 2 per cent (four of 218 fractures) for all 186 patients who completed the study as well as for the ninety-nine patients in Group 1 who completed the study (two of 118 fractures) and for the eighty-seven patients in Group 2 who completed the study (two of 100 fractures). All four fractures that were complicated by infection were located in the distal half of the tibia. The neurovascular status of these extremities remained normal throughout the period of the study. The infections were diagnosed between three and four weeks after the injury. Only one patient had osteomyelitis. The other three had gross purulence extending to the bone but needed only minimum débridement. All four fracture sites were infected with Staphylococcus aureus, and one was infected with Group-D gamma-hemolytic Streptococcus also. The organisms were sensitive to both the orally administered and the intravenously administered antibiotics. All four fractures were treated with formal irrigation and débridement on several occasions as well as with extended intravenous therapy with antibiotics. One needed external fixation. None of the fractures needed coverage with a flap. The seventy tibial fractures were evenly divided between Group 1 and Group 2. Seventeen fractures in Group 1 and eighteen fractures in Group 2 were in the proximal half of the tibia. Eighteen fractures in Group 1 and seventeen fractures in Group 2 were in the distal half of the tibia. The rate of infection in the distal half of the tibia was 11 per cent (four of thirty-five fractures). The fourteen fractures in the twelve patients who chose not to participate in the study healed without infection after treatment with the Group-1 protocol. The twenty-two fractures that were classified as a cortical violation (a non-displaced fracture that involved a single cortex of the bone, as seen on plain radiographs) healed in an average of six weeks (range, three to ten weeks). The remaining fractures healed in an average of fourteen weeks (range, five to thirty-nine weeks).
Three of the most important factors that physicians who treat fractures caused by low-velocity gunshots must consider are management of the wound, appropriate treatment of the fracture, and antibiotic therapy9,13,16,17,22. The extent of tissue damage inflicted by a bullet is influenced by the bullet's velocity, weight, size, shape, and tumbling characteristics as well as by the character of the tissue that it strikes17. The capacity for wounding increases exponentially as velocity increases (KE = 1/2 mv2, where KE is kinetic energy and mv2 is mass times velocity squared). Bullets from most low-velocity weapons used in a civilian setting travel at an average of 1000 to 2000 feet (305 to 610 meters) per second while bullets from high-velocity weapons exceed 2000 feet (610 meters) per second. Most military ammunition travels at 2400 to 2900 feet (732 to 884 meters) per second and that from M-16 rifles travels at 3250 feet (991 meters) per second17. It is only when the impact velocity is greater than 600 feet (183 meters) per second that shock waves from the missile have much effect. Temporary cavitation becomes apparent when velocity exceeds 1000 feet (305 meters) per second and becomes extensive when velocity exceeds 2000 feet (610 meters) per second16. The size of the area of cavitation is directly proportional to the energy transmitted34. Bone can fracture when struck by a missile with a velocity of only 200 feet (sixty-one meters) per second17. Damage to long bones depends, to some extent, on the area of involvement. When a low-velocity bullet strikes the cancellous portion of the metaphysis of a bone, a so-called drill-hole defect frequently results with little or no comminution. Conversely, when the same bullet strikes the shaft, it often produces a comminuted fracture with so-called butterfly fragments at right angles to the wound track7,8. Nerves and tendons are usually pushed out of the way secondary to their mobility, and the vessels, being elastic, often give and stretch with low-velocity-gunshot injuries29. Thoresby and Darlow30 showed that bullets are not sterilized by firing. They also proved that organisms can be drawn into the wound by temporary cavitation. Contamination from bits of clothing and skin are also likely sources of infection30,33. Although our study of the role of antibiotic therapy was randomized and prospective, it had several limitations. The hypothesis that no antibiotics are needed for the routine care of low-velocity-gunshot fractures was not addressed. It may be that the rate of infection would be no different. We reviewed the literature to compare the rate of infections associated with fractures caused by low-velocity gunshots and treated with various antibiotic regimens (Table II). As the groups were relatively small and there were small differences between them, we performed statistical analysis with use of the Fisher exact text26. For the present study, the exact algorithm p value was equal to 1.00—that is, there was no significant difference between the two groups. The p values in several studies9,34 were less than 10 per cent, with use of chi-square analysis. The significance of the results is questionable because of the small number of subjects in those studies.
The design of our study was biased against the inclusion of some injuries that are routinely stabilized and in favor of some injuries that are not. For example, most fractures of the femoral shaft are treated with intramedullary nailing and thus were excluded from the study. However, most fractures of the humeral shaft are treated non-operatively and were included. The femoral fractures that were included in this study were non-displaced so-called cracks, cortical violations, or metaphyseal drill-hole injuries. They were treated with above-the-knee casts initially, followed by cast-bracing or fracture-bracing (if necessary). All of the patients who had a femoral fracture were managed with protected weight-bearing with use of crutches. Although the design of the study was biased, we believe that the impact of the bias was minor. We are not attempting to offer new methods for fixation of fractures; we simply wanted to determine if orally administered antibiotics could be used effectively in the treatment of fractures that do not need operative fixation. Our study had a high rate of follow-up because, in addition to the senior ones of us (T. P. K., M. J. P., and J. L.), we had three full-time research fellows (P. R. S., K. A., and R. B. R.) who worked on the project. While in the hospital, each patient was questioned about personal data and was interviewed daily. Often, the real name and address of the patient were not revealed until just before discharge. Information obtained from the patient included his or her driver's license number and Social Security number as well as the name, address, and telephone number (home and work) of a friend or relative. Each patient was contacted weekly by telephone or mail, or both, to ascertain if he or she had moved and to provide reminders about follow-up appointments at the clinic. If necessary, transportation to and from the clinic was arranged. The easiest patients to follow were those who were incarcerated. A telephone call was made to the authorities at the prison, and the patient was returned for a follow-up examination. The weekly contact to verify if the vital information on the patient had changed was the single most important factor in our high rate of follow-up. We concluded that patients who have an extra-articular fracture of a long bone, other than one involving the distal half of the tibia, from a low-velocity gunshot can be managed with oral antibiotic therapy without an increased prevalence of infection compared with that noted with intravenous antibiotic therapy. Our data suggest that fractures caused by a low-velocity gunshot in the distal half of the tibia are associated with a higher rate of infection regardless of the antibiotic regimen. Additional studies are needed to evaluate whether fractures caused by a low-velocity gunshot in the distal half of the tibia are more susceptible to infection, as our findings suggest.
NOTE: The authors thank Leslie Bernstein, Ph.D., Department of Biostatistics, University of Southern California School of Medicine, for her assistance in the design of the study and in the statistical analysis of the data.
*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?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |