TIME UNTIL THE START OF ANTIBIOTIC PROPHYLAXIS AND THE RISK OF OPEN FRACTURE INFECTION: A SYSTEMATIC REVIEW

Marchiori, João Guilherme Tavares; Nunes, Ana Paula Ferreira

doi:10.1590/1413-785220243202e263176

Acessibilidade / Reportar erro

Brasil

Acta Ortopédica Brasileira

Español English

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Systematic Review Article/Orthopedic Trauma • Acta Ortop Bras 32 (2) • 2024 • https://doi.org/10.1590/1413-785220243202e263176 copy

TIME UNTIL THE START OF ANTIBIOTIC PROPHYLAXIS AND THE RISK OF OPEN FRACTURE INFECTION: A SYSTEMATIC REVIEW

TEMPO ATÉ O INÍCIO DA ANTIBIOTICOPROFILAXIA E O RISCO DE INFECÇÃO EM FRATURAS EXPOSTAS: UMA REVISÃO SISTEMÁTICA

Authorship SCIMAGO INSTITUTIONS RANKINGS

ABSTRACT

Open fractures are highly incident injuries closely related to the modern life, in which accidents caused by motor vehicles or other machines impart high energy to bone tissue. Individual morbidity is represented by the functional impairment resultant of infection, nonunion, or vicious healing. In terms of public health, there are huge costs involved with the treatment of these fractures, particularly with their complications. One of the critical issues in managing open fractures is the use of antibiotics (ATB), including decisions about which specific agents to administer, duration of use, and ideal timing of the first prophylactic dose. Although recent guidelines have recommended starting antibiotic prophylaxis as soon as possible, such a recommendation appears to stem from insufficient evidence. In light of this, we conducted a systematic review, including studies that addressed the impact of the time to first antibiotic and the risk of infectious outcomes. Fourteen studies were selected, of which only four found that the early initiation of treatment with antibiotics is able to prevent infection. All studies had important risks of bias. The results indicate that this question remains open, and further prospective and methodologically sound studies are necessary in order to guide practices and health policies related to this matter. Level of Evidence II; Therapeutic Studies Investigating the Results Level of Treatment.

Keywords:
Antibacterial Agents; Fractures, Bone; Infection Control

Study Year of publication Design	Inclusion	Sample data Timing to first ATB	Outcome definition	Risk of bias	Study risk of bias
Dellinger et al. 1988¹⁵15 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(11):1320-7. doi:10.1001/archsurg.1988.01400350034004. https://doi.org/10.1001/archsurg.1988.01... Retrospective cohort	Femur, humerus, leg bones, forearm bones All classifications Multicentric > 14y No comorbidities	N= 240 (263 fx) Minimum follow-up 21d Gt I: 25% Gt II: 47% Gt IIIA: 19% Gt IIIB: 5% Gt IIIC: 5% Method of counting time to first ATB undefined	Clinical criteria	A=M B=L C=S D=L E=S F=S G=M	S
Patzakis et al. 1989¹⁶16 Patzakis MJ, Wilkins J. Factors influencing infection rate in open fracture wounds. Clin Orthop Relat Res. 1989;(243):36-40. Prospective cohort	Any age Any bone	N= 1.104 or 1.390? Undefined follow-up Method of counting time to first ATB undefined	Clinical criteria, confirmed by microbiology	A=S B=S C=NI D=L E=C F=M G=M	C
Al-Arabi et al. 2008¹⁷17 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(8):900-5. doi:10.1016/j.injury.2007.02.043. Erratum in: Injury. 2008;39(3):381. https://doi.org/10.1016/j.injury.2007.02... Prospective cohort	Femur, humerus, leg bones, forearm bones All classifications	N=133 Undefined follow-up Method of counting time to first ATB undefined	Clinical criteria (edema, erythema, discharge, pain), cultures when possible	A=C B=C C=S D=L E=C F=S G=M	C
Enninghorst et. al 2011¹⁸18 Enninghorst N, McDougall D, Hunt JJ, Balogh ZJ. Open tibia fractures: timely debridement leaves injury severity as the only determinant of poor outcome. J Trauma. 2011;70(2):352-7. Prospective cohort	> 18y, trauma center, all classifications of open diaphyseal tíbia fx	N=89 Gt I: 25% Gt II: 30% Gt IIIA: 20% Gt IIIB: 24% Gt IIIC: 1% Follow-up: 12m	Surgical debridement indication or long term systemic ATB	A=M B=L C=S D=L E=NI F=S G=M	S
Thomas et al. 2013¹⁹19 Thomas SH, Arthur AO, Howard Z, Shear ML, Kadzielski JL, Vrahas MS. Helicopter emergency medical services crew administration of antibiotics for open fractures. Air Med J. 2013;32(2):74-9. doi:10.1016/j.amj.2012.06.007. https://doi.org/10.1016/j.amj.2012.06.00... Prospective cohort	Any age, extremity open fractures	N= 138 Follow-up: 6m 60 patients: ATB pre-hospital (helicopter) 78 patients: ATB hospital Method of counting time to first ATB: time of admission and time of trauma	Composite Endpoint (superficial or deep infection or nonunion)	A=C B=L C=S D=L E=C F=C G=M	C
Leonidou et al. 2014²⁰20 Leonidou A, Kiraly Z, Gality H, Apperley S, Vanstone S, Woods DA. The effect of the timing of antibiotics and surgical treatment on infection rates in open long-bone fractures: a 6-year prospective study after a change in policy. Strategies Trauma Limb Reconstr. 2014;9(3):167-71. doi:10.1007/s11751-014-0208-9. https://doi.org/10.1007/s11751-014-0208-... Prospective cohort	Open long bones fractures	N= 212 (220 fx) Analysis for first ATB included 139 patients Follow-up: until bone healing or a procedure for nonunion or infection Gt I: 36,6% Gt II: 19,9% Gt IIIA: 24,8% Gt IIIB: 18,6% Method of counting time to first ATB: time of admission and time of trauma	Purulent discharge from deep fascia, dehiscence; "radiological evidence" or cultures	A=C B=L C=S D=L E= C F=S G=M	C
Weber et al. 2014²¹21 Weber D, Dulai SK, Bergman J, Buckley R, Beaupre LA. Time to initial operative treatment following open fracture does not impact development of deep infection: a prospective study of 736 subjects. J Orthop Trauma. 2014;28(11):613-9. Prospective cohort	Long bones open fx of adults. All classifications	N=686 (737 fx) Gt I: 29% Gt II: 37% Gt IIIA: 21% Gt IIIB: 12% Gt IIIC: 1% Follow-up: 90d or phone interview at 12m	Surgical debridement indication or long term systemic ATB	A=M B=L C=S D=L E=S F=S G=M	S
Zumsteg et al. 2014²²22 Zumsteg JW, Molina CS, Lee DH, Pappas ND. Factors influencing infection rates after open fractures of the radius and/or ulna. J Hand Surg Am. 2014;39(5):956-61. doi:10.1016/j.jhsa.2014.02.008. https://doi.org/10.1016/j.jhsa.2014.02.0... Retrospective cohort	18y Radius and/or ulna open fx	N=200 Variable follow-up (max 6m) Gt I: 22% Gt II: 24% Gt III: 55% Data from medical records	Deep infection as an indication or surgical debridement, assessed from medical records or phone calls	A=S B=C C=S D=L E=S F=C G=M	C
Lack et al. 2015²³23 Lack WD, Karunakar MA, Angerame MR, Seymour RB, Sims S, Kellam JF, et al. Type III open tibia fractures: immediate antibiotic prophylaxis minimizes infection. J Orthop Trauma. 2015;29(1):1-6. doi:10.1097/BOT.0000000000000262. Erratum in: J Orthop Trauma. 2015;29(6):e213. https://doi.org/10.1097/BOT.000000000000... Retrospective cohort	Type III open tibia fractures	N=137 Follow-up 90d Gt IIIA: 52% Gt IIIB/IIIC: 48% Method of counting time to first ATB: time of admission and time of trauma	CDC	A=M B=L C=S D=L E=L F=S G=M	S
Johnson et al 2017.²⁴24 Johnson JP, Goodman AD, Haag AM, Hayda RA. Decreased Time to Antibiotic Prophylaxis for Open Fractures at a Level One Trauma Center. J Orthop Trauma. 2017;31(11):596-9. doi:10.1097/BOT.0000000000000928. https://doi.org/10.1097/BOT.000000000000... Cross-sectional	> 18y Limb and axial bones All classifications Data from medical records	N=100 1 group N= 50 before early ATB protocol. 1 group N=50 after protocol institution Undefined follow-up	Surgery indication	A=S B=C C=S D=L E=S F=S G=M	C
Assunção ALF, Oliveira de ST. 2020²⁵25 Assunção ALF, Oliveira de ST. Clinical Audit of Primary Treatment of Open Fractures: Antibiotic Treatment and Tetanus Prophylaxis. Rev Bras Ortop (Sao Paulo). 2020;55(3):284-92. doi:10.1055/s-0039-3402470. https://doi.org/10.1055/s-0039-3402470... Prospective cohort	> 18y, trauma center, data from medical records.	N=241 Gt I: 20% Gt II: 19% Gt III: 21,6% NC: 39,4% Time from admission to first ATB	NS	A=C B=C C=S D=C E=M F=S G=M	C
Hendrickson et al 2020²⁶26 Hendrickson SA, Donovan R, Harries L, Wright TC, Whitehouse MR, Khan U. Time to intravenous antibiotic administration (TIbiA) in severe open tibial fractures: Impact of change to national guidance. Injury. 2020;51(4):1086-90. doi:10.1016/j.injury.2020.03.005. https://doi.org/10.1016/j.injury.2020.03... Retrospective cohort	Type IIIB open tíbia fx	N= 156 (159 fx) Minimum follow-up 1y Median 26 m (IQR 18-39) Method of counting time to first ATB: time of trauma	Deep infection confirmed by microbiology	A=M B=L C=L D=L E=L F=M G=M	M
Roddy et al. 2020²⁷27 Roddy E, Patterson JT, Kandemir U. Delay of Antibiotic Administration Greater than 2 Hours Predicts Surgical Site Infection in Open Fractures. Injury. 2020;51(9):1999-2003. doi:10.1016/j.injury.2020.04.031. https://doi.org/10.1016/j.injury.2020.04... Retrospective cohort	Upper and lower limb open fx, all classifications, data from medical records	N= 230 Minimum follow-up: 30d, endpoint assessment at 90d	CDC NHSN	A=M B=L C=S D=L E=S F=S G=M	S
Zuelzer et al. 2021²⁸28 Zuelzer DA, Hayes CB, Hautala GS, Akbar A, Mayer RR, Jacobs CA, et al. Early Antibiotic Administration Is Associated with a Reduced Infection Risk When Combined with Primary Wound Closure in Patients with Open Tibia Fractures. Clin Orthop Relat Res. 2021;479(3):613-9. doi:10.1097/CORR.0000000000001507. https://doi.org/10.1097/CORR.00000000000... Retrospective cohort	> 18y, trauma center, data from medical records, rescue sheets, Gustilo I, II, IIIA	N=127 Gt I: 27,6% Gt II: 48,8% Gt IIIA: 23,6% Minimum follow-up: 6w	CDC	A=M B=L C=S D=L E=M F=S G=M	S

Study	Analysis	Results	Comments	Earl ATB: recommendation x usual practice
Dellinger et. al. 1988¹⁵15 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(11):1320-7. doi:10.1001/archsurg.1988.01400350034004. https://doi.org/10.1001/archsurg.1988.01...	Chi-square Fisher Student's t Kaplan-Meier Logistic regression	Time to 1°ATB < 3h: 16% infected; > 3h: 17% infected. p=0,9784 Mean time to 1º ATB in infected: 2,0h (+-1,1h); non-infected: 2,2h (+-1,4h)	Method of counting the time to first ATB not informed 22% lost to follow-up at 6m	No mention about recommendation or practice
Patzakis et al. 1989¹⁶16 Patzakis MJ, Wilkins J. Factors influencing infection rate in open fracture wounds. Clin Orthop Relat Res. 1989;(243):36-40.	Chi-square	Time to first ATB <3h (364 fx): 4,7% infected. . >3h (661 fx): 7,4% infected p= 0,087 (Yates 0,114)	No information on follow-up No control for confounding variables Method of counting the time to first ATB not informed. Dichotomization of time to first ATB variable. No information on time as a continuous variable Divergence regarding composition of the cohort (1.104 ou 1.390?) No apparent distinction between superficial and deep infection	Recommends ATB as soon as possible after lesion
Al-Arabi et al. 2007¹⁷17 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(8):900-5. doi:10.1016/j.injury.2007.02.043. Erratum in: Injury. 2008;39(3):381. https://doi.org/10.1016/j.injury.2007.02...	Fisher Linear Regression	Time to first ATB < 6h: 5,7% infected > 6h: 22,2% infected p=0,1144	No control for confounding variables No information regarding central tendency measures for follow-up Method of counting the time to first ATB not informed A non-specified number of more severe fx (IIIB and IIIC) lost to follow-up, with no information on their basal characteristics 80% statistical power for a reduction of 10% in infection rate	No mention about recommendation or practice
Enninghorst et al. 2011¹⁸18 Enninghorst N, McDougall D, Hunt JJ, Balogh ZJ. Open tibia fractures: timely debridement leaves injury severity as the only determinant of poor outcome. J Trauma. 2011;70(2):352-7.	Means Student's t Mann-Whitney U Chi-square Univariate, bivariate, multiple regression	Cohort mean: 1,2h (+-0,3h) Incidence of infection: 16,8% No difference in time to first ATB between infected and non-infected	Indefinition regarding classification of intervention and outcome No missing data information	No mention about recommendation or practice
Thomas et al. 2013¹⁹19 Thomas SH, Arthur AO, Howard Z, Shear ML, Kadzielski JL, Vrahas MS. Helicopter emergency medical services crew administration of antibiotics for open fractures. Air Med J. 2013;32(2):74-9. doi:10.1016/j.amj.2012.06.007. https://doi.org/10.1016/j.amj.2012.06.00...	Fisher Chi-square Kruskal-Wallis	Pre-hospital ATB group: 60 patients (13 completed follow-up) 1 outcome (infection or nonunion [7,7%]) Hospital ATB group: 78 patients. (70 completed follow-up) 9 outcomes nonunion [12,9%]) P=1,0 60,2% lost to follow-up	No control for confounding variables Inconsistencies in classification of intervention, without proper control (potentially affects internal validity) High losses to follow-up Meticulous statistical analysis and discussion about limitations	No mention about recommendation or practice
Leonidou et al. 2014²⁰20 Leonidou A, Kiraly Z, Gality H, Apperley S, Vanstone S, Woods DA. The effect of the timing of antibiotics and surgical treatment on infection rates in open long-bone fractures: a 6-year prospective study after a change in policy. Strategies Trauma Limb Reconstr. 2014;9(3):167-71. doi:10.1007/s11751-014-0208-9. https://doi.org/10.1007/s11751-014-0208-...	Fisher	Time to 1°ATB < 3h: 14% infected; > 3h: 12,5% infected. p=1,0	No control for confounding variables No information regarding central tendency measures for follow-up 39,6% lost to follow-up Inconsistencies in classification of intervention, without proper control (potentially affects internal validity) Inconsistencies in information of sample composition and in records of losses	Usual practice: ATB in less than 3 hours from lesion
Weber et al. 2014²¹21 Weber D, Dulai SK, Bergman J, Buckley R, Beaupre LA. Time to initial operative treatment following open fracture does not impact development of deep infection: a prospective study of 736 subjects. J Orthop Trauma. 2014;28(11):613-9.	Medians Mann-Whitney U Simple and multiple regression	6% of infection Median to 1° ATB among infected: 2h37min. Median to first ATB among non-infected: 3h5min p=0,67 Logistic regression: OR 1,0 (IC95% 0,95-1,05)	Sound methodology Method of counting the time to first ATB not informed Few losses to follow-up. Intervention not known in 15% of patients No definite conclusion on the association of early ATB and infection, as most patients received late ATB	Usual practice
Zumsteg et al. 2014²²22 Zumsteg JW, Molina CS, Lee DH, Pappas ND. Factors influencing infection rates after open fractures of the radius and/or ulna. J Hand Surg Am. 2014;39(5):956-61. doi:10.1016/j.jhsa.2014.02.008. https://doi.org/10.1016/j.jhsa.2014.02.0...	Wilcoxon Fisher Chi-square Logistic Regression	32% lost to follow-up, with no information on their basal characteristics Mean time to 1° ATB: 1,6 +− 0,9h among infected; 2,6 +− 2,2 horas among noninfected ATB < 3h: 159 patients (6% infected). ATB > 3h: 41 patients (2% infected p=0,40 10 infections (5%), on average 118 days after first stabilization	Many confounders not controlled Inconsistencies in classification of intervention High losses to follow-up Upper limb open fractures have less risk of getting infected, so big samples may be needed to investigate such associations.	No mention about recommendation or practice
Lack et al. 2015²³23 Lack WD, Karunakar MA, Angerame MR, Seymour RB, Sims S, Kellam JF, et al. Type III open tibia fractures: immediate antibiotic prophylaxis minimizes infection. J Orthop Trauma. 2015;29(1):1-6. doi:10.1097/BOT.0000000000000262. Erratum in: J Orthop Trauma. 2015;29(6):e213. https://doi.org/10.1097/BOT.000000000000...	Chi-square Student's t Logistic Regression	Time to 1°ATB < 66min: 7% infected; > 66min: 25% infected p=0,0063 ROC: 66min (AUC=0,63 p=0,03) Logistic regression: ATB > 66min: OR = 3,78 (CI95% 1,26-14,11 p= -0,016)	Sound methodology and analysis Gives a cut-off time to first ATB Sample calculation for a power of 80% Late ATB is a independent predictor of infection Inconsistencies in classification of intervention, without proper control (potentially affects internal validity)	Recommends ATB as soon as possible, preferably at pre-hospital level
Johnson et al. 2017²⁴24 Johnson JP, Goodman AD, Haag AM, Hayda RA. Decreased Time to Antibiotic Prophylaxis for Open Fractures at a Level One Trauma Center. J Orthop Trauma. 2017;31(11):596-9. doi:10.1097/BOT.0000000000000928. https://doi.org/10.1097/BOT.000000000000...	Chi-square Mann Whitney U Student's t	Time to first ATB dropped from 123,1min to 35,7min (p=0,0003). Incidence of infection = 10% for both groups	Time to first ATB counted from admission time (risk of bias due to classification of intervention) Outcome defined as indication of surgery (not precise and subjective) Follow-up not defined Small sample (few outcomes, low power)	Usual practice: first ATB as soon as possible from admission
Assunção ALF, Oliveira de ST. 2020²⁵25 Assunção ALF, Oliveira de ST. Clinical Audit of Primary Treatment of Open Fractures: Antibiotic Treatment and Tetanus Prophylaxis. Rev Bras Ortop (Sao Paulo). 2020;55(3):284-92. doi:10.1055/s-0039-3402470. https://doi.org/10.1055/s-0039-3402470...	Frequencies Chi-square	Infection: ≤ 3h from admission: 15,7% > 3 h from admission: 26,1% p = 0.0350	Confounders and co-interventions not listed	Usual practice (preoperative ATB)
Hendrickson et al. 2020²⁶26 Hendrickson SA, Donovan R, Harries L, Wright TC, Whitehouse MR, Khan U. Time to intravenous antibiotic administration (TIbiA) in severe open tibial fractures: Impact of change to national guidance. Injury. 2020;51(4):1086-90. doi:10.1016/j.injury.2020.03.005. https://doi.org/10.1016/j.injury.2020.03...	Medians and IQR Logistic Regression	Time to 1° ATB: median 162 min (IQR: 120-207) Time to 1° ATB x Infection (regression analysis): Continuous: p=0,431 1h: p=0,099 3h: p=0,848	Sound methodology and analysis Main confounders accounted for, including multicollinearity tests Outcome assessed with objective criteria Potential risk of beta error, as most patients took late ATB (>2h)	Usual practice: early/pre-hospital ATB
Roddy et al. 2020²⁷27 Roddy E, Patterson JT, Kandemir U. Delay of Antibiotic Administration Greater than 2 Hours Predicts Surgical Site Infection in Open Fractures. Injury. 2020;51(9):1999-2003. doi:10.1016/j.injury.2020.04.031. https://doi.org/10.1016/j.injury.2020.04...	Chi-square Mann Whitney U ROC Cox regression	Deep infection: 6% Median to 1° ATB in infected: 83min Median to 1° ATB non-infected: 61min p=0,053 Cut-off 120min ROC (AUC 0.62, 95% CI [0.50 – 0.75], p = 0.042) OR 2,4 [CI95% 1,1-5,7] p=0,036.	Sound methodology and analysis Gives a cut-off time to first ATB CI of AUC do not show a definitive benefit of cut-off found Small sample (low power) 130 patients missed (29%) e 78 with no information on time to first ATB (17%)	Recommends ATB as soon as possible
Zuelzer et al. 2021²⁸28 Zuelzer DA, Hayes CB, Hautala GS, Akbar A, Mayer RR, Jacobs CA, et al. Early Antibiotic Administration Is Associated with a Reduced Infection Risk When Combined with Primary Wound Closure in Patients with Open Tibia Fractures. Clin Orthop Relat Res. 2021;479(3):613-9. doi:10.1097/CORR.0000000000001507. https://doi.org/10.1097/CORR.00000000000...	Chi-square Fisher ANOVA Binary regression Logistic regression ROC curve	Infection: ≤ 150 from admission: 3% > 150 from admission: 20% Odds Ratio 5.6 [95% CI 1.4 to 22.2]; p = 0.01	Sound methodology, detection bias risk, risk of bias due to classification of intervention (non-standardized sources of data)	ATB as soon as possible after lesion (practice and recommendation)

ATHA EDITORA Rua: Machado Bittencourt, 190, 4º andar - Vila Mariana - São Paulo Capital - CEP 04044-000, Telefone: 55-11-5087-9502 - São Paulo - SP - Brazil
E-mail: actaortopedicabrasileira@uol.com.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] Correspondence: João Guilherme Tavares Marchiori. 60/1102, Taciano Abaurre Street, Enseada do Suá, Vitória, ES, Brasil. 29050470. joaogtm@yahoo.com.br