Trauma after Transplant: Hold the Antibiotics Please

Methods

For the purpose of our study, we reviewed retrospectively a trauma registry at our Level 1 trauma center. In this review, we evaluated all TP from January 2006 through December 2010 and included those who had documented solid-organ transplants. We then compared these patients with their concurrent NTP counterparts, using 1:3 propensity matching.

We also compared the two groups for other comorbidities than trauma and for lengths of stay and deaths. In addition, we evaluated the groups' infection rates and antimicrobial treatment. We designated a unit that we named antibiotic-days, which we defined as the sum of days on which a patient was treated with a particular antibiotic. Penicillins and first- or second-generation cephalosporins, tetracyclines, metronidazole, vancomycin, and linezolid were defined as narrow-spectrum antibiotics, and beta-lactam/beta-lactam–inhibitor combinations, carbapenems, fluoroquinolones, third- or fourth-generation cephalosporins, trimethoprim/sulfamethoxazole, and aminoglycosides as broad-spectrum antibiotics. The Wilcoxon rank-sum test and the Fisher exact test were used for comparisons, with the exception of antibiotic-day comparisons, which were analyzed with the use of unpaired t-tests. The study was approved by the institutional review board of Virginia Commonwealth University.

Results

During the time frame of the study, there were 14,140 trauma team admissions, of which 17 followed a solid-organ transplant, for an incidence of 0.12%. The 17 patients in whom these transplants were done included 13 who had kidney, two who had liver, one who had a kidney/liver, and one who had a kidney/pancreas transplant. These patients were compared with 51 NTP who were propensity matched for injury severity score (ISS), age, and gender. The average ISS was 8.5±9.3 TP versus 11.0±9.8 NTP, p=0.24; the average age in years was 49.6±14.7 TP versus 48.9±14.7 NTP, p=0.67; and the percentage of male patients in each group was 76% (13/17) TP versus 71% (36/51) NTP, p=0.76.

The two groups were also comparable in their types of trauma, with most suffering blunt trauma (82.4% [14/17] TP vs. 90% [46/51] NTP, p=0.5). The groups had similar initial Glasgow Coma Scale (GCS) scores (13.2±4.5 TP vs. 13.9±2.5 NTP, p=0.6), serum lactate concentrations (2.0±1.8 mmol/L TP vs. 2.3±1.5 mmol/L NTP, p=0.39), and base deficits (–1.5±4.0 TP vs. 0.6±3.0 NTP, p=0.21). Comorbidities were more common in the TP than in the NTP group, as shown in Table 1.

No differences were found in outcome variables. The two study groups had similar lengths of stay (days) for ventilator use: 0.1±0.3 TP versus 0.4±1.6 NTP, p=0.9; duration of stay in the intensive care unit (ICU): 0.2±0.6 TP versus 2.4±5.9 NTP, p=0.16; and duration of hospitalization: 5.2±6.8 TP versus 7.5±10.2 NTP, p=0.86. Two deaths occurred in each group (p=0.26). Despite similar complications and infections, a greater proportion of transplant patients received antimicrobial agents when compared to the NTP group (59% vs. 39%, p=0.17), as shown in Table 2, although this did not reach statistical significance. Additionally, there was a trend for the TP group to receive antimicrobials for a longer duration (8.35 vs. 3.9 days, p=0.17, Table 2). There were also no differences in the mean number of antibiotic days between the transplant recipients and the matched control group (Table 2) when the spectra of the antibiotic drugs used for the two groups were separated into narrow- and broad-spectrum drugs.

Discussion

Solid-organ transplants continue to increase, with more of the patients receiving them surviving for longer periods with their transplanted organs [6,7]. For the vast majority of such patients this success results from the use of immune suppression, with the primary consequence of this paradigm being an increased risk of infection with poor outcomes when infection is not treated aggressively [8,9]. The ever-present risk of infection commonly influences clinical practice when these patients are admitted to hospitals for reasons other than transplant-related complications. Transplant recipients interact with the health-care system frequently, both before and after receiving their transplants. This augmented interaction, together with transplant recipients' immunocompromised state, often results in a real and perceived threat of infections with multidrug-resistant pathogens such as methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), and Pseudomonas aeruginosa.

The empiric use of antimicrobial drugs in transplant recipients is often influenced by the clinical perception of risk from these pathogens, and leads to both appropriate and inappropriate use of broad-spectrum antimicrobial drugs in this patient population. The initiation of treatment with empiric antimicrobial drugs to “protect” vulnerable transplant recipients may be justified if there is a clear danger of infection related to the acute condition of the patient [8]. Trauma is certainly a circumstance in which a greater risk of infection is present; trauma results in damage to important protective barriers, and the hospital-acquired infections associated with this have detrimental consequences [10,11].

We were concerned that TP admitted to our hospital system for trauma-related injuries would have a higher incidence of infections than contemporary NTP, with detrimental short-term outcomes. The retrospective study described here demonstrated that TP did not have a higher incidence of infection than patients without solid-organ transplants and the immune suppression associated with organ transplantation, despite having similar ISS scores. Transplant patients also had clinical outcomes similar to those of NTP with respect to hospital length of stay, ventilator days, and ICU length of stay. There was a trend toward an increased duration of antimicrobial treatment during the trauma-related hospital stay for TP as compared with NTP, suggesting that the clinicians treating the TP did perceive an increased risk of infection in the population of TP. However, we did not find a difference in exposure to antibiotic agents in the transplant and control groups in our study when the antibacterial agents used were divided into narrow- and broad-spectrum drugs. This observation makes it unlikely that the low rate of infection among traumatized transplant recipients in our study population was a result of the routine use of broad-spectrum antibacterial drugs in this group. However, given the sample size, it is impossible to determine the effect of the trend toward greater use of antimicrobial agents among the transplant recipients in our study on the rates of infection in this population.

Barone et al. conducted a retrospective study of 12 traumatized transplant patients (seven kidney, five kidney/pancreas) identified over a 40-month period at the University of Arkansas [3]. Although they identified infectious complications among this group of patients, they did not include a contemporary control group of TP without solid-organ transplants, making it difficult to interpret the relative significance of the infections they observed. The most common complications among their cohort were closed fractures, which constitute an outcome on which we did not focus in our case-control study.

Our single-center study has some limitations that should be considered when interpreting the data in this paper. The sample size is small and does not represent the full spectrum of solid-organ–transplant types. This finding may have influenced the risk of infection because not all types of organ transplants are treated in the same way with respect to the degree of immune suppression of the patient. Different types of organ transplant are also associated with other comorbid conditions that could influence the clinical outcomes that we found in our patient population (e.g., lung TP and ventilator days). Moreover, our patient population as a whole was mildly injured, with an average ISS score well below 15 in both of the patient groups in the study. These findings may not reflect results for patients who are more severely injured.

Although the outcomes of immunocompromised patients who suffer trauma have been evaluated [12–14], clinical outcomes specific for different types of infection in traumatized solid-organ–transplant recipients have not been reported in the published English literature. A PubMed search of “organ transplant AND trauma AND infection” (limits: human, English, clinical trial) disclosed no studies addressing this aspect of the clinical care of TP. We believe that our study is the first to specifically investigate the incidence of infections in traumatized solid-organ–transplant recipients, and our results indicate that there is no indication for the increased use of empiric or prophylactic antimicrobial therapy in these patients solely on the basis of a perceived risk of infection.

When equally matched for degree of injury, TP and NTP have similar trauma-related outcomes. There also appears to be no difference in infectious complications in these two groups of patients, yet a greater number of the transplant patients in our study had exposure to more days of antibiotic administration. Similar protocols should apply to both groups, to avoid the overuse of antimicrobial drugs and ensure the maintenance of microbial susceptibility patterns.