Abstract
Background:
Existing guidelines recommend emergency surgical debridement of open fractures within six hours after injury. The aim of this study was to systematically review the association between time to operative debridement of open fractures and infection.
Methods:
Searches of the MEDLINE, EMBASE, and Cochrane computerized literature databases and manual searches of bibliographies were performed. Randomized controlled trials and cohort studies (retrospective and prospective) evaluating the association between time to operative debridement and infection after open fractures were included. Descriptive and quantitative data were extracted. A meta-analysis of patient cohorts who underwent early or delayed debridement was performed with use of a random effects model.
Results:
The initial search identified 885 references. Of the 173 articles inspected further on the basis of the title, sixteen (six prospective and ten retrospective cohort studies with a total of 3539 open fractures) were included. No significant difference in the infection rate was detected between open fractures debrided early or late according to any of the time thresholds used in the included studies. Sensitivity analyses demonstrated no difference in infection rate between early and late debridement in subgroups defined according to the Gustilo-Anderson classification, level of evidence, depth of infection, or anatomic location.
Conclusions:
The data did not indicate an association between delayed debridement and higher infection rates when all infections were considered, when only deep infections were considered, or when only more severe open fracture injuries were considered. On the basis of this analysis, the historical “six-hour rule” has little support in the available literature. It is important to realize that additional carefully conducted studies are needed and that elective delay of treatment of patients with open fractures is not recommended.
Level of Evidence:
Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.
Until 150 years ago, open fractures were synonymous with sepsis and death, necessitating immediate amputation as the definitive treatment1. Advances in antimicrobial therapy, fracture stabilization, and wound management dramatically decreased mortality from open fractures although the number of open fractures and similar high-energy injuries has increased1,2. Epidemiologic studies have shown that open long-bone fractures occur at a rate of 11.5 per 100,000 persons per year3,4. The prevalence of infection following internal fixation of fractures is approximately 5% overall but may exceed 30% in open fractures5. Musculoskeletal infections place a cost burden on total health care expenditures, with the reported lifetime cost of the most severe open fracture injuries being as high as $680,0006. Traditional clinical guidelines suggest treatment of open fractures with an initial operative debridement within six hours after injury to reduce the risk of infection. It is believed that the “six-hour rule” originated from a study conducted on guinea pigs by Friedrich in 18987. He found that when debridement of open wounds was performed within six hours, all animals remained healthy7. In 1973, Robson et al. quantified wound bacterial counts to define an “open fracture infection threshold,” characterized as a density of ≥105 organisms per gram of tissue, and found that this threshold was reached within a mean of 5.17 hours after injury8.
Although expedient and appropriate treatment of these severe injuries should be the goal, there are circumstances in which delaying the initial debridement may benefit or at the very least not harm the patient. In a large observational study, it was noted that initial debridement of 42% of open tibial fractures was delayed for more than six hours9. The consequences of this delay in treatment are unknown. The purpose of the present systematic review and meta-analysis was to evaluate the association between the time to initial operative debridement of open fractures and the development of infectious complications.
Data Sources
Two of the authors (M.L.S. and S.Y.) independently carried out a comprehensive search of the MEDLINE, EMBASE, and Cochrane computerized literature databases (through December 3, 2010) for randomized controlled trials, quasi-randomized controlled trials, and cohort studies (both prospective and retrospective) that evaluated the effect of early compared with late debridement of open fractures on infection outcomes.
The medical subject headings (MeSH terms) used were “open fracture” or “open fractures” and “debridement.” Reviewers traced the bibliographies of the retrieved articles, including review articles, for citations missed by the electronic search. The senior investigators (S.M. and J.A.) also reviewed their personal files and associated bibliographies for additional citations.
Study Selection
Two of the authors (M.L.S. and S.Y.) reviewed the abstract titles for relevance and determined which articles potentially contained relevant information. If an article was deemed eligible by either reviewer, the abstract was retrieved and reviewed in full. Only studies published in English were reviewed. Studies were included if they described (1) a minimum of twenty-six subjects; (2) data for patients over the age of eighteen years; (3) clinical and radiographic evidence of fracture union at the time of follow-up; (4) evidence of the completeness of wound-healing; (5) infectious outcomes; (6) open long-bone fractures (femur, tibia, and/or humerus) with time to debridement as a metric; and (7) a Level-I, II, or III therapeutic or prognostic study design. Studies were excluded if they (1) did not meet the above inclusion criteria; (2) were not performed on human subjects; (3) did not allow outcomes of open fractures to be distinguished from outcomes of closed fractures; (4) classified gunshot wounds as open fractures10; or (5) did not involve long bones (e.g., were fingers or toes). The review of pertinent abstracts was performed by three of the authors (M.L.S., S.Y., and K.D.B.). If any abstract was deemed relevant by any reviewer, the full text of the article was reviewed by the same three reviewers. If two of the three reviewers felt that the article should be kept, it was included in the review. Overall agreement among the reviewers was 64%, and the free marginal kappa was 0.5311 (indicating moderate agreement)12.
Data Extraction
Two of the authors (M.L.S. and S.Y.) independently extracted data, including general information (author and year of publication), type of study, period of patient enrollment, mean age, sex distribution, mean duration of follow-up, definition of infection (e.g., osteomyelitis or cellulitis), infection rate, time to initial debridement, Gustilo-Anderson classification, fracture location, type and timing of antibiotic administration, wound management strategy, and fracture management strategy.
Three of the authors (M.L.S., S.Y., and K.D.B.) assessed the methodological quality of the studies according to the criteria of Zaza et al.13, who described a systematic method of assessing the quality of observational studies of preventive medicine interventions. This quality assessment method spans five major areas of study design: descriptions of the population and intervention, sampling, measurement, data analysis, and interpretation of results. No summary score is generated by this tool13. The reviewers resolved disagreement by discussion and consensus.
Data Synthesis
A meta-analysis using a DerSimonian and Laird random effects model14-16 was then performed to compare dichotomous outcomes between early and delayed debridement groups aggregated from the parent studies. In cases in which no events occurred in the parent study, a continuity correction of 0.5 was added to each cell to permit analysis, as described in the Cochrane handbook17,18.
In addition to examining the relationship between the primary outcome of any infectious complication and the exposure factor of delayed compared with early debridement, five sensitivity analyses were performed. The first sensitivity analysis also evaluated the outcome of any infection, but it compared the outcomes of early and late debridement on the basis of each of the time thresholds provided by the authors of the parent studies (five19, six20-29, eight30,31, and twelve32 hours). The second sensitivity analysis evaluated only studies that described “deep infections” (defined in the parent studies as infections extending below the fascia26,33, as purulent discharge or osteomyelitis20,25,30,31, or as clinical diagnosis of pain and/or erythema and/or discharge with positive wound cultures27 (see Appendix). The third sensitivity analysis evaluated the outcome of any infection according to the level of evidence of the study, to determine whether focusing on studies with a higher level of evidence could uncover any significant differences in infection rate between the early and late debridement cohorts that were not identified in the primary meta-analysis. In this secondary analysis, the Level-II19,24,27,32-34 studies were analyzed separately from the Level-III20-23,25,26,28-31 studies. The fourth sensitivity analysis evaluated the results of the studies on the basis of the severity of the injury35, with Gustilo-Anderson type-I and II open fractures analyzed separately from type-III open fractures20-22,28,29. The final sensitivity analysis separately evaluated the results of the studies that included only lower-extremity fractures19-23,25,26,28,29,31 and those that included only tibial fractures21-23,25,28,29,31.
Forest plots were generated to qualitatively assess study heterogeneity and to provide summary estimates. A funnel plot and the Eggers intercept method were used to assess the existence of publication bias due to small-study effects. Because of the heterogeneity among the studies, which included different mixtures of patient types and injury severities, we utilized a random effects model (DerSimonian and Laird) to provide a conservative method of combining the effects of multiple studies15,16.
Source of Funding
No external funding sources were utilized in this investigation.
The initial search yielded 885 citations: 294 from MEDLINE, 576 from EMBASE, and fifteen from the Cochrane Review (Fig. 1). Of these, 712 articles were excluded on the basis of the title because they clearly represented a review paper, represented an editorial or contained commentary without primary data, or were unrelated to our topic. Of the remaining 173 articles, 144 were excluded on the basis of the abstract because they failed to satisfy the predetermined inclusion criteria or because they were editorial in nature or represented a review article, case report, or erratum. Nineteen of the remaining twenty-nine articles were excluded when the full article was reviewed because it failed to meet the inclusion criteria. This left ten unique studies from our initial review, and a manual reference search revealed six additional studies. Our systematic review included these sixteen articles with a total of 3539 open fractures.
The funnel plot to assess study heterogeneity was relatively symmetric, with no perceivable publication bias. The Eggers intercept was 0.85 (95% confidence interval [CI], −0.66 to 2.36; p = 0.24).
Study Characteristics
The Appendix summarizes the key characteristics of the included studies. Six prospective studies provided Level-II evidence19,24,27,32-34 and ten retrospective studies provided Level-III evidence20-23,25,26,28-31. No study was randomized on the basis of the time to debridement.
The time threshold used for the comparison between the early and late debridement groups was six hours in nine studies20,22-29, five hours in two19,21, eight hours in two30,31, twelve hours in one32, and not specifically reported in two33,34. When raw data were provided and permitted separation between early and late debridement on the basis of a six-hour time threshold, the data were extracted and incorporated into the meta-analysis according to this time threshold21.
Five studies used a cohort with upper and lower-extremity fractures24,27,30,32,33, and eleven studies were limited to lower-extremity fractures19-23,25,26,28,29,31,34. Seven of the latter studies evaluated only open tibial fractures21-23,25,28,29,31, one examined only femoral fractures26, and one classified the fracture types as tibial or non-tibial32. The authors of this last study reported a significantly higher infection rate for tibial compared with non-tibial fractures32. Of the studies that included both upper and lower-extremity fractures, both Harley et al. and Dellinger et al. reported that lower-extremity fractures were associated with a higher rate of deep infection30,33.
We performed a quality analysis of each study according to the method described by Zaza et al. for assessing preventive medicine studies13. The population was well described in thirteen of the sixteen studies, one study did not describe consistently which bone was involved in the open injury32, and two studies had very little demographic information regarding the cohort28,31. Thirteen of the sixteen studies described the intervention adequately; the antibiotic administration was not clear in the remaining three studies19,25,28. One study did not fully describe why patients were excluded; it was mentioned that some patients were transferred, but the reason was not specified24. Seven of the sixteen studies did not use the full population over the entire study period19,21,24,26,27,31,34. One study was performed in Nigeria, with initial and definitive care differing from that in the more developed world34. An open fracture with exposed bone was described in all studies. The performance of a reliability analysis for either the classification of bone exposure (according to the Gustilo-Anderson type) or the diagnosis of infection was not described in any study. The criteria for deep infection varied among the studies and ranged from osteomyelitis30,31 to cellulitis20, with only a few studies using culture data to confirm the presence of infection22,24,27,30,32. There was no consistent definition of superficial infection among the studies. Five studies adjusted for potential confounding factors with use of multivariate logistic regression19,25,26,31,33, and one study showed no difference in multiple confounding factors between groups30. No study corrected for the use of multiple tests. Three studies had <80% follow-up22,24,33. Six studies either did not report the time to follow-up or had inaccuracies in their reporting25-29,32.
Effect of Delayed Debridement on Overall Infection Rates
Fourteen studies provided early and late debridement times and infection rates; these studies included 3217 open fractures and a total of 396 infections suitable for meta-analysis. On further review, one study32 provided data for a large number of patients but did not adequately define the fracture population. Therefore, separate analyses were performed with and without inclusion of that study. Infection was defined differently in each study (see Appendix), ranging from positive intraoperative and wound cultures to culture-positive chronic osteomyelitis (more than four weeks) and nonunion. Seven studies presented the deep infection rate only21,23,24,26,29-31, seven presented a combined rate for deep and superficial infections20,22,27,28,32-34, and two presented separate data for combined deep and superficial infections and for deep infections alone22,25.
The overall infection rates ranged from 4% to 63%. No significant difference in the overall infection rate between early and late debridement was detected (Fig. 2). The weighted cumulative odds ratio (OR) of developing an infection after late compared with early debridement was 0.91 (95% CI, 0.70 to 1.18). The risk difference between the early and late groups was −1% (95% CI, −4% to 2%) in favor of late debridement, although this difference was not significant (p = 0.46). The odds ratio was unchanged with inclusion of the single article with the heterogeneous fracture population32 (OR, 0.93; 95% CI, 0.74 to 1.17). Two studies indicated that the time to debridement was a significant factor in increasing infection outcomes21,34; however, one of these articles provided insufficient data for inclusion in the meta-analysis34, and further analysis of the other article21 with use of six rather than five hours as the threshold between early and late debridement yielded an odds ratio for infection of 3.68 (95% CI, 0.96 to 14.06), which did not reach significance.
When the studies were analyzed according to the time thresholds for early and late debridement used by the primary authors, no significant difference in infection rates was detected with use of any of the following cutoffs: five hours (OR, 0.96; 95% CI, 0.54 to 1.71; p = 0.88), six hours (OR, 0.81; 95% CI, 0.53 to 1.24; p = 0.34), eight hours (OR, 1.15; 95% CI, 0.51 to 2.59; p = 0.73), or twelve hours (OR, 1.04; 95% CI, 0.62 to 1.73; p = 0.789).
Effect of Depth of Infection
When only deep infections were considered, no significant difference in the infection rate between early and late debridement was detected (Fig. 3). The weighted cumulative odds ratio of infection was 1.07 (95% CI, 0.74 to 1.54). The risk difference between the early and delayed groups was 1% (95% CI, −2% to 4%); this difference was not significant (p = 0.69).
Effect of Injury Severity and Delayed Debridement
Five studies reported injury severity according to the Gustilo-Anderson classification36 and evaluated the effects of delayed debridement on infection rates. For Gustilo-Anderson type-I and II fractures, 310 patients with an overall infection rate of 8% in four of these studies were available for analysis. The infection rate was 12% in the early debridement group and 5% in the late debridement group. The risk difference between the early and delayed groups was −4% (95% CI, −10% to 2%) in favor of late debridement, although this difference was not significant (p = 0.25) (Fig. 4). The weighted cumulative odds ratio of developing an infection after late debridement was 0.58 (95% CI, 0.25 to 1.33).
For Gustilo-Anderson type-III fractures, 276 patients with an overall infection rate of 12.7% were available for analysis. The infection rate was 15% in the early debridement group and 11% in the late debridement group. The risk difference between the early and delayed groups was −4% (95% CI, −12% to 5%) in favor of late debridement, although this difference was not significant (p = 0.44) (Fig. 5). The weighted cumulative odds ratio of developing an infection after late debridement was 0.84 (95% CI, 0.31 to 2.31).
The Effect of Study Level of Evidence
When only deep infections were evaluated according to the level of evidence, the weighted cumulative odds ratio of infection after late compared with early debridement did not differ significantly between the studies that provided Level-II evidence19,24 and Level-III21,23,25,26,29-31 evidence. For studies with Level-II evidence, the weighted cumulative odds ratio of infection was 1.13 (95% CI, 0.63 to 2.03) for late debridement. For studies with Level-III evidence, the weighted cumulative odds ratio of infection was 1.04 (95% CI, 0.65 to 1.65) for late debridement.
The Effect of Anatomic Location of the Fracture (Lower Extremity Only and Tibia Only)
When infection rates were evaluated according to anatomic location, the weighted cumulative odds ratio of infection after late compared with early debridement was not significantly different for lower-extremity fractures or for only tibial fractures compared with all fractures. The weighted cumulative odds ratio of infection after late debridement of lower-extremity fractures was 0.88 (95% CI, 0.62 to 1.26). The weighted cumulative odds ratio of infection after late debridement of tibial fractures was 0.89 (95% CI, 0.5 to 1.57).
In this review, we present the aggregation and analysis of sixteen systematically identified studies on the effect of late debridement of open fractures on the infection rate. The meta-analysis revealed no association between later debridement times and higher infection rates when all infections were considered, when only deep infections were considered, or when only more severe open fracture injuries were considered.
Strengths and Weaknesses of This Review
In this review, we attempted to extract as much data as possible from the individual studies and we performed a systematic analysis of study quality with use of a previously described methodological tool13. Fourteen studies (3217 fractures) that included the time to operative debridement as a recorded metric were available for the meta-analysis. Furthermore, the study data were amenable to subgroup analysis according to fracture severity (Gustilo-Anderson type), infection depth, study level of evidence, and anatomic location.
Inclusion of retrospective cohort studies has inherent risks of bias, confounding, and associations that are not improved by aggregating studies. However, despite their limitations in methodology, such studies included a substantial number of patients, and ignoring them might have affected the external validity of the findings of the meta-analysis. Furthermore, differing infection definitions, wound handling, irrigation practices, antibiotic administration, patient comorbidities, virulence of potential contaminants, injury characteristics, and skeletal instability could not be controlled for in this analysis, and these will require further study.
In addition, the details of antibiotic administration were not well described in most of the studies (see Appendix). As antibiotic use is likely a major factor in reducing infection rates, it is an important factor to consider when designing future studies. Two studies reported neither the type nor the timing of antibiotic administration25,28. Nine additional studies reported the type of antibiotic administered but did not describe the timing after injury20-23,26,29-31,34. Dellinger et al. reported a mean time to antibiotic administration of 2.1 hours (range, 0.2 to nine hours)33 but did not make any associations between delayed antibiotic administration and the infection rate. Spencer et al. reported that all patients received antibiotics within four hours of injury24. Interestingly, Patzakis and Wilkins noted a higher infection rate in patients who received antibiotics more than three hours after injury (7.4% compared with 4.7%)32. Similarly, Pollak et al. noted that a prolonged time between injury and hospitalization (more than two hours)—which served as a proxy for the timing of antibiotic administration—was associated with a higher rate of infection19. Finally, Al-Arabi et al. did not find an association between the timing of antibiotic administration and the infection rate, but they did note that two patients who had delays in both operative debridement and antibiotic administration developed infections27.
Other Studies—Clinical Data
Two studies included in the systematic review33,34 did not provide extractable data for the meta-analysis. Dellinger et al.33 performed a prospective study that evaluated the development of deep and superficial infections in a cohort of 263 upper and lower-extremity fractures. The authors determined that the mean time to debridement was 5.7 hours for patients who did not develop infections compared with 5.0 hours for patients without infection; this difference was not significant. Ikem et al.34 prospectively evaluated a series of sixty-three consecutive open fractures and noted that patients who developed infections had a significantly longer delay to initial debridement. Of note, the clinical practice of orthopaedics in Nigeria differs to a substantial extent compared with that in more developed nations. In that series, fracture stabilization was achieved with skeletal traction, Steinmann pins, external fixation, and plaster casting with a cast window cut out for wound care.
Two additional retrospective case series were excluded from the systematic review and meta-analysis on the basis of the inclusion criterion involving the minimum level of evidence37,38. Furthermore, two Level-III studies of pediatric patients were excluded from the systematic review and meta-analysis on the basis of the inclusion criterion involving patient adulthood39,40. Nonetheless, the results of these additional clinical studies were consistent with the findings of our meta-analysis, suggesting that early initial debridement may not be a critical factor in reducing infection rates following open fractures.
Other Studies—Experimental Data
In a 1961 study involving subcutaneous inoculation of guinea pigs, Burke concluded that prevention of infection was best achieved when antibiotics were administered prior to bacterial inoculation and that the effect of systemic antibiotics decreased as the time interval after inoculation increased, reaching a threshold of no effect after three hours41. In another study involving administration of cephradine to prevent tibial osteomyelitis in a rabbit model, administration of the antibiotic prior to inoculation was shown to be significantly more effective than administration after inoculation; however, administration up to four hours after injury still had a dramatic effect on the prevention of infection42. There is also substantial clinical evidence to support early antibiotic administration as a critical factor in preventing infection after open fractures43,44.
Implications of Our Review
Although initial expedient and appropriate irrigation and debridement of open fracture injuries should be the goal, there are circumstances in which early debridement may not be possible. For example, hospital facilities in a rural setting or in a remote military theater may not have the resources to accommodate treatment of these complex injuries39,45,46. Under these circumstances, immediate transfer to a specialty hospital may offer the patient an improved injury outcome even if it results in delayed initial debridement39,46. Moreover, many of these open fractures present outside normal hospital working hours. Given the limited resources (including surgical assistants, experienced operating room staff, and appropriate equipment and imaging services) available outside normal working hours in some hospital settings, some investigators have suggested an increased risk of complications associated with after-hours surgery47-51. Therefore, although urgent management of open fractures is encouraged, it has been suggested that abiding by the historical “six-hour rule” may offer a disservice to patients when antibiotic administration and appropriate initial fracture care are provided under suboptimal operating conditions.
In this review, late surgical debridement was not associated with a higher infection rate in patients with open fractures. Even patients with severe injuries, classified as Gustilo-Anderson type-III fractures, did not have a higher infection rate with late initial debridement. Given the available data, it is difficult to determine the length of time between injury, administration of antibiotics, and operative debridement that provides the best outcome for the patient with an open fracture. Ideally, any delay should be minimized if possible. It is important to realize that additional carefully conducted studies are needed and that purposeful delay of treatment of patients with open fractures is not recommended.
Unanswered Questions and Future Directions
Our study cannot be considered to conclusively invalidate the “six-hour rule.” However, it provides sufficient equipoise to justify prospective investigations into the timing of initial debridement of open fractures. Clearly, open fractures can lead to substantial infectious morbidity. The identification and analysis of modifiable risk factors (including the previously suggested factors of time between injury and admission to a trauma center19, quality of debridement52, timing of antibiotic administration32, substantial bone loss [>2 cm] 19, fracture location30,32-34, patient comorbidities, and smoking status53-57) in well-designed prospective trials will allow us to decrease the morbidity associated with open fractures.
Tables showing the characteristics and outcomes of the included studies are available with the online version of this article as a data supplement at jbjs.org.
Browner
BD;
Jupiter
JB;
Levine
AM;
Trafton
PG;
Krettek
C, . Skeletal trauma: basic science, management, and reconstruction. 4th ed. Philadelphia, PA: W.B. Saunders Company; 2008.
DeLee
JC;
Stiehl
JB. Open tibia fracture with compartment syndrome. Clin Orthop Relat Res.
1981;160:175-84.[PubMed]
Court-Brown
CM;
Rimmer
S;
Prakash
U;
McQueen
MM. The epidemiology of open long bone fractures. Injury.
1998;29:529-34.[PubMed][CrossRef]
Howard
M;
Court-Brown
CM. Epidemiology and management of open fractures of the lower limb. Br J Hosp Med.
1997;57:582-7.[PubMed]
Trampuz
A;
Widmer
AF. Infections associated with orthopedic implants. Curr Opin Infect Dis.
2006;19:349-56.[PubMed][CrossRef]
Chung
KC;
Saddawi-Konefka
D;
Haase
SC;
Kaul
G. A cost-utility analysis of amputation versus salvage for Gustilo type IIIB and IIIC open tibial fractures. Plast Reconstr Surg.
2009;124:1965-73.[PubMed][CrossRef]
Friedrich
PL. Die aseptische Versorgung frischer Wunden, unter Mittheilung von Thier-Versuchen uber die Auskeimungszeit von Infectionserregern in frischen Wunden. Archiv fur Klinsche Chirugie.
1898:288-310.
Robson
MC;
Duke
WF;
Krizek
TJ. Rapid bacterial screening in the treatment of civilian wounds. J Surg Res.
1973;14:426-30.[PubMed][CrossRef]
Namdari
S;
Baldwin
KD;
Matuszewski
P;
Esterhai
JL;
Mehta
S. Delay in surgical débridement of open tibia fractures: an analysis of national practice trends. J Orthop Trauma.
2011;25:140-4.[PubMed][CrossRef]
Volgas
DA;
Stannard
JP;
Alonso
JE. Current orthopaedic treatment of ballistic injuries. Injury.
2005;36:380-6.[PubMed][CrossRef]
Randolph
JJ. Free-marginal multirater kappa: an alternative to Fleiss’ fixed-marginal multirater kappa. Joensuu, Finland: Joensuu University; 2005.
Landis
JR;
Koch
GG. The measurement of observer agreement for categorical data. Biometrics.
1977;33:159-74.[PubMed][CrossRef]
Zaza
S;
Wright-De Agüero
LK;
Briss
PA;
Truman
BI;
Hopkins
DP;
Hennessy
MH;
Sosin
DM;
Anderson
L;
Carande-Kulis
VG;
Teutsch
SM;
Pappaioanou
M. Data collection instrument and procedure for systematic reviews in the Guide to Community Preventive Services. Task Force on Community Preventive Services. Am J Prev Med.
2000;18(
1 Suppl):44-74.[PubMed][CrossRef]
DerSimonian
R;
Laird
N. Meta-analysis in clinical trials. Control Clin Trials.
1986;7:177-88.[PubMed][CrossRef]
Bax
L;
Yu
LM;
Ikeda
N;
Tsuruta
H;
Moons
KG. Development and validation of MIX: comprehensive free software for meta-analysis of causal research data. BMC Med Res Methodol.
2006;6:50.[PubMed][CrossRef]
MIX: comprehensive free software for meta-analysis of causal research data. Version 1.7. 2008. .
The Cochrane Collaboration Cochrane handbook for systematic reviews of interventions. 2005. .
The Cochrane Collaboration. RevMan 5.1 user guide. 2011. .
Pollak
AN;
Jones
AL;
Castillo
RC;
Bosse
MJ;
MacKenzie
EJ; LEAP Study Group. The relationship between time to surgical debridement and incidence of infection after open high-energy lower extremity trauma. J Bone Joint Surg Am.
2010;92:7-15.[PubMed][CrossRef]
Bednar
DA;
Parikh
J. Effect of time delay from injury to primary management on the incidence of deep infection after open fractures of the lower extremities caused by blunt trauma in adults. J Orthop Trauma.
1993;7:532-5.[PubMed][CrossRef]
Kindsfater
K;
Jonassen
EA. Osteomyelitis in grade II and III open tibia fractures with late debridement. J Orthop Trauma.
1995;9:121-7.[PubMed][CrossRef]
Khatod
M;
Botte
MJ;
Hoyt
DB;
Meyer
RS;
Smith
JM;
Akeson
WH. Outcomes in open tibia fractures: relationship between delay in treatment and infection. J Trauma.
2003;55:949-54.[PubMed][CrossRef]
Ashford
RU;
Mehta
JA;
Cripps
R. Delayed presentation is no barrier to satisfactory outcome in the management of open tibial fractures. Injury.
2004;35:411-6.[PubMed][CrossRef]
Spencer
J;
Smith
A;
Woods
D. The effect of time delay on infection in open long-bone fractures: a 5-year prospective audit from a district general hospital. Ann R Coll Surg Engl.
2004;86:108-12.[PubMed][CrossRef]
Charalambous
CP;
Siddique
I;
Zenios
M;
Roberts
S;
Samarji
R;
Paul
A;
Hirst
P. Early versus delayed surgical treatment of open tibial fractures: effect on the rates of infection and need of secondary surgical procedures to promote bone union. Injury.
2005;36:656-61.[PubMed][CrossRef]
Noumi
T;
Yokoyama
K;
Ohtsuka
H;
Nakamura
K;
Itoman
M. Intramedullary nailing for open fractures of the femoral shaft: evaluation of contributing factors on deep infection and nonunion using multivariate analysis. Injury.
2005;36:1085-93.[PubMed][CrossRef]
Al-Arabi
YB;
Nader
M;
Hamidian-Jahromi
AR;
Woods
DA. The effect of the timing of antibiotics and surgical treatment on infection rates in open long-bone fractures: a 9-year prospective study from a district general hospital. Injury.
2007;38:900-5.[PubMed][CrossRef]
Sungaran
J;
Harris
I;
Mourad
M. The effect of time to theatre on infection rate for open tibia fractures. ANZ J Surg.
2007;77:886-8.[PubMed][CrossRef]
Tripuraneni
K;
Ganga
S;
Quinn
R;
Gehlert
R. The effect of time delay to surgical debridement of open tibia shaft fractures on infection rate. Orthopedics.
2008;31.
Harley
BJ;
Beaupre
LA;
Jones
CA;
Dulai
SK;
Weber
DW. The effect of time to definitive treatment on the rate of nonunion and infection in open fractures. J Orthop Trauma.
2002;16:484-90.[PubMed][CrossRef]
Reuss
BL;
Cole
JD. Effect of delayed treatment on open tibial shaft fractures. Am J Orthop (Belle Mead NJ).
2007;36:215-20.[PubMed]
Patzakis
MJ;
Wilkins
J. Factors influencing infection rate in open fracture wounds. Clin Orthop Relat Res.
1989;243:36-40.[PubMed]
Dellinger
EP;
Miller
SD;
Wertz
MJ;
Grypma
M;
Droppert
B;
Anderson
PA. Risk of infection after open fracture of the arm or leg. Arch Surg.
1988;123:1320-7.[PubMed][CrossRef]
Ikem
IC;
Oginni
LM;
Bamgboye
EA. Open fractures of the lower limb in Nigeria. Int Orthop.
2001;25:386-8.[PubMed][CrossRef]
Gustilo
RB;
Anderson
JT. Prevention of infection in the treatment of one thousand and twenty-five open fractures of long bones: retrospective and prospective analyses. J Bone Joint Surg Am.
1976;58:453-8.[PubMed]
Gustilo
RB. Management of open fractures. An analysis of 673 cases. Minn Med.
1971;54:185-9.[PubMed]
Yang
EC;
Eisler
J. Treatment of isolated type I open fractures: is emergent operative debridement necessary?Clin Orthop Relat Res.
2003;410:289-94.[PubMed][CrossRef]
Naique
SB;
Pearse
M;
Nanchahal
J. Management of severe open tibial fractures: the need for combined orthopaedic and plastic surgical treatment in specialist centres. J Bone Joint Surg Br.
2006;88:351-7.[PubMed][CrossRef]
Skaggs
DL;
Kautz
SM;
Kay
RM;
Tolo
VT. Effect of delay of surgical treatment on rate of infection in open fractures in children. J Pediatr Orthop.
2000;20:19-22.[PubMed][CrossRef]
Skaggs
DL;
Friend
L;
Alman
B;
Chambers
HG;
Schmitz
M;
Leake
B;
Kay
RM;
Flynn
JM. The effect of surgical delay on acute infection following 554 open fractures in children. J Bone Joint Surg Am.
2005;87:8-12.[PubMed][CrossRef]
Burke
JF. The effective period of preventive antibiotic action in experimental incisions and dermal lesions. Surgery.
1961;50:161-8.[PubMed]
Worlock
P;
Slack
R;
Harvey
L;
Mawhinney
R. The prevention of infection in open fractures. An experimental study of the effect of antibiotic therapy. J Bone Joint Surg Am.
1988;70:1341-7.[PubMed]
Gosselin
RA;
Roberts
I;
Gillespie
WJ. Antibiotics for preventing infection in open limb fractures. Cochrane Database Syst Rev.
2004;1:.
Patzakis
MJ;
Harvey
JP
Jr;
Ivler
D. The role of antibiotics in the management of open fractures. J Bone Joint Surg Am.
1974;56:532-41.[PubMed]
Townley
WA;
Nguyen
DQ;
Rooker
JC;
Dickson
JK;
Goroszeniuk
DZ;
Khan
MS;
Camp
D. Management of open tibial fractures - a regional experience. Ann R Coll Surg Engl.
2010;92:693-6.[PubMed][CrossRef]
Gopal
S;
Majumder
S;
Batchelor
AG;
Knight
SL;
De Boer
P;
Smith
RM. Fix and flap: the radical orthopaedic and plastic treatment of severe open fractures of the tibia. J Bone Joint Surg Br.
2000;82:959-66.[PubMed][CrossRef]
Henriques
JP;
Haasdijk
AP;
Zijlstra
F; Zwolle Myocardial Infarction Study Group. Outcome of primary angioplasty for acute myocardial infarction during routine duty hours versus during off-hours. J Am Coll Cardiol.
2003;41:2138-42.[PubMed][CrossRef]
Taffinder
NJ;
McManus
IC;
Gul
Y;
Russell
RC;
Darzi
A. Effect of sleep deprivation on surgeons’ dexterity on laparoscopy simulator. Lancet.
1998;352:1191.[PubMed][CrossRef]
Ricci
WM;
Gallagher
B;
Brandt
A;
Schwappach
J;
Tucker
M;
Leighton
R. Is after-hours orthopaedic surgery associated with adverse outcomes? A prospective comparative study. J Bone Joint Surg Am.
2009;91:2067-72.[PubMed][CrossRef]
Gawande
AA;
Zinner
MJ;
Studdert
DM;
Brennan
TA. Analysis of errors reported by surgeons at three teaching hospitals. Surgery.
2003;133:614-21.[PubMed][CrossRef]
Saleem
MA;
Kannam
H;
Aronow
WS;
Weiss
MB;
Kalapatapu
K;
Pucillo
AL;
Monsen
CE. The effects of off-normal hours, age, and gender for coronary angioplasty on hospital mortality in patients undergoing coronary angioplasty for acute myocardial infarction. Am J Cardiol.
2004;93:763-4.[PubMed][CrossRef]
Merritt
K. Factors increasing the risk of infection in patients with open fractures. J Trauma.
1988;28:823-7.[PubMed][CrossRef]
Bowen
TR;
Widmaier
JC. Host classification predicts infection after open fracture. Clin Orthop Relat Res.
2005;433:205-11.[PubMed][CrossRef]
Castillo
RC;
Bosse
MJ;
MacKenzie
EJ;
Patterson
BM; LEAP Study Group. Impact of smoking on fracture healing and risk of complications in limb-threatening open tibia fractures. J Orthop Trauma.
2005;19:151-7.[PubMed][CrossRef]
Harvey
EJ;
Agel
J;
Selznick
HS;
Chapman
JR;
Henley
MB. Deleterious effect of smoking on healing of open tibia-shaft fractures. Am J Orthop (Belle Mead NJ).
2002;31:518-21.[PubMed]
White
CB;
Turner
NS;
Lee
GC;
Haidukewych
GJ. Open ankle fractures in patients with diabetes mellitus. Clin Orthop Relat Res.
2003;414:37-44.[PubMed][CrossRef]
Harrison
WJ;
Lewis
CP;
Lavy
CB. Open fractures of the tibia in HIV positive patients: a prospective controlled single-blind study. Injury.
2004;35:852-6.[PubMed][CrossRef]