a b s t r a c t

Introduction: Civilian gunshot open-fracture injuries portray a significant health burden to patients. Use of anti- biotics is endorsed by guideline recommendations for the prevention of post-traumatic infections, however, an- timicrobial selection and their associated outcomes remains unclear. Therefore, we sought to compare infectious and other clinical outcomes between three antimicrobial cohorts in patients with gunshot-related fractures re- quiring operative intervention.

Materials and methods: Patients were identified by retrospectively querying the University of Kentucky Trauma Registry for gunshot wound victims. A narrow regimen, an expanded gram-negative regimen, and a regimen containing a fluoroquinolone antimicrobial were identified for comparison. The primary outcome was a compos- ite of infections at or before 14 days of hospitalization. Secondary endpoints included hospital length of stay, in- cidence of multidrug resistant bacteria and methicillin-resistant Staphylococcus aureus colonization, number of drug-related adverse events, number of Clostridium Difficile infections, and 30-day mortality.

Results: 252 patients were selected for inclusion: 126 in the narrow regimen, 49 in the expanded gram-negative regimen, and 77 in the fluoroquinolone-based regimen. There were no statistical differences in the primary end- point of early infectious outcomes between groups (p = 0.1797). The expanded gram-negative regimen was as- sociated with increased hospital length of stay, and increased incidence of multi-drug resistant bacteria and methicillin-resistant Staphylococcus aureus colonization. There were no statistically significant differences in any of the remaining secondary endpoints.

Conclusion: In this study evaluating civilian gunshot trauma, broad spectrum antibiotic coverage was not associ- ated with improvements in post-traumatic infections. A randomized trial is needed to confirm these results.

(C) 2019

Introduction

Firearm injuries represent a significant healthcare burden in the United States with an estimated 60,000-80,000 Americans affected each year [1,2]. orthopedic injuries sustained in these incidents consti- tute a major infectious risk due to wound exposure to the environment. Although local wound care, practice of aseptic technique, and surgery represented most of the medical innovation prior to the 1900’s, it was not until World War II that the development of penicillin helped clini- cians revolutionize open-fracture management [3,4]. When employed in a timely manner, antibiotics help reduce the incidence of post-

* Corresponding author.

E-mail address: [email protected] (J.A. Woolum).

traumatic infection, subsequently decreasing patient morbidity and mortality [5-7].

Gunshot wound open-fractures are a unique subset of bony injury that possess many uncertainties regarding their appropriate manage- ment. Specifically, recommendations on the use of antimicrobial agents remains vague, with evidence for their use stemming from several dated studies containing small patient populations, or evaluating mili- tary ballistic trauma rather than civilian trauma [5,6,8]. Furthermore, guidance on antimicrobial use and selection are nearly absent in major clinical guidelines, with differing opinions represented in each docu- ment. Although most experts agree that coverage of gram-positive or- ganisms is necessary, the need for inclusion of an expanded gram- negative antimicrobial spectrum remain unknown for civilian gunshot trauma [9-13]. This question is especially important given the increas- ing amounts of evidence associating broad-spectrum antimicrobial use

https://doi.org/10.1016/j.ajem.2019.158358

0735-6757/(C) 2019

with collateral damage in the forms of Antimicrobial resistance, adverse effects, and poorer wound healing.

Due to the uncertainty surrounding the most appropriate antimicro- bial regimen for gunshot wound fractures, we sought to evaluate a larger number of civilian patients suffering from gunshot wound frac- ture injuries to determine if expanded gram-negative antimicrobial cov- erage offers any benefit over a narrower-spectrum antimicrobial regimen.

Methods

Study design and setting

We conducted a single-center retrospective cohort study in an aca- demic medical center emergency department (ED). The University of Kentucky Chandler Medical Center (UKCMC) is a tertiary referral hospi- tal and level one trauma center, servicing N100,000 patients within the ED annually. The study was conducted after approval by the health system’s institutional review board and was performed in accordance with the Strength of the Reporting of Observational Studies in Epidemi- ology guidelines (Supplementary Appendix Table 1).

Study participants

Study participants were identified by retrospectively querying the University of Kentucky Trauma Registry (UKTR). The UKTR is a clinical document used to record all pertinent injuries, vital signs, and labora- tory values during trauma patient triage. After querying patient encoun- ters between January 1, 1999, and January 1, 2017, subjects meeting predetermined criteria were included for analysis. Enrollees were sepa- rated into three antimicrobial groups depending on the antibiotic they received within the first 48 h of their stay. The three cohorts were la- beled as being narrow-spectrum, expanded gram-negative spectrum (EGN), and fluoroquinolone-based (FB) (see Section 2.4 Definitions and outcomes).

Inclusion criteria and exclusion criteria

Patients presenting to the UKED with open-fracture injuries second- ary to a gunshot injury and requiring operative repair were initially identified for study inclusion. Identification was accomplished by que- rying the UKTR with the cause-code phrase, “gunshot wound”, and “GSW”. Individuals with an Abbreviated Injury Scale 2005 code containing injuries to the abdomen, thorax, and head were excluded from the study [14]. Additional exclusion criteria included failure to re- ceive operative intervention, and lack of antibiotic administration within 48 h of hospital presentation. Patients meeting all appropriate in- clusion and exclusion criteria were included in the final analysis.

Definitions and outcomes

During the study period, antibiotic administration for gunshot- related fractures was non-protocolized and was dictated at the treating clinician’s discretion. Study participants were divided into three sepa- rate cohorts depending on the antimicrobial regimen they received dur- ing their first 48 h of hospital stay. The narrow-spectrum group included individuals with orders for either cefazolin or clindamycin monother- apy. The EGN group included patients with an antimicrobial regimen containing either an extended-spectrum beta-lactam, or an aminogly- coside. Lastly, the FB group included patients with orders for a fluoro- quinolone agent, with or without additional antimicrobial agents.

The primary outcome was a composite endpoint of post-traumatic infections, up to day 14 of hospitalization. Patients discharged before day 14 were included in the analysis. Infections comprising the compos- ite endpoint included any documented bacteremia, skin and soft tissue infection (SSTI), or osteomyelitis.

Secondary outcomes included hospital length of stay, incidence of positive multidrug resistant (MDR) bacteria and methicillin-resistant Staphylococcus aureus (MRSA) surveillance cultures identified during hospitalization, number of documented drug-related adverse effects, in- cidence of hyperkalemia, incidence of acute renal failure, number of Clostridium difficile infections, and 30-day mortality.

Data collection and variables

After patient enrollment, necessary identifiers, demographics and clinical laboratory values were collected and managed using the Re- search Electronic Data Capture (REDCap) tools hosted at The University of Kentucky [15]. Investigators then utilized the electronic health record to obtain pertinent medications administered during the patient’s stay and detect other pre-specified outcomes of the study.

Incidence of post-traumatic infections and the secondary outcomes

of positive MDR and MRSA surveillance cultures were captured via a ret- rospective database review of patient charts for their respective labora- tory results. Drug-related adverse events, hyperkalemia, Clostridium difficile infections, and acute renal failure, were each identified via a search for their associated ICD-9 and ICD-10 codes (Supplementary Ap- pendix Table 2).

Data analysis

Comparisons were made between the three antimicrobial regimen cohorts for all pre-specified endpoints. Categorical variables were assessed using chi-square and Fisher’s exact test, as appropriate. Contin- uous variables were tested for normality using the Shapiro-Wilk nor- mality test and histograms. Normally distributed variables were reported using means and standard deviations (SD) and compared using ANOVA’s. Otherwise, they were reported using medians and 1st and 3rd quartiles (Q1-Q3) and compared using Kruskal-Wallis test. All analyses were conducted in R programming language, version 3.5.1 (R Foundation for Statistical Computing, Vienna, Austria). Statistical signif- icance was set at a p-value of b0.05.

Results

Study cohort

After querying the UKTR, 966 were identified during the pre-defined study period of which 714 patients were then excluded (Fig. 1). The most common reasons for exclusion included lack of antibiotic adminis- tration times (n = 324) and failure to receive antibiotic therapy (n = 133). This yielded an overall sample size of 252 patients available for analysis. Of the 252 patients, 126 were placed into the narrow regimen cohort, 49 were relegated to the EGN cohort, and 77 were clustered into the FB cohort.

Baseline demographics for each cohort were well matched and may be found in Table 1. Patients were predominantly young, adult males with injuries tending to occur in a lower extremity. injury severity scores (ISS) were the same for each group. There was some statistically significant variation occurring between each antibiotic cohort. Antibi- otic selection for each cohort is demonstrated in Fig. 2.

Primary outcome

For the primary outcome of post-traumatic bacteremia, SSTI, and os- teomyelitis, we identified three infections among the study population for an infection rate of 1.2%. We were unable to detect any significant differences in infection rates between any of the three cohorts (p = 0.1797).

Of the three infections identified, two occurred in the expanded gram-negative cohort, and one occurred in the narrow spectrum cohort (Table 2). Infections occurring in the expanded gram-negative regimen

Fig. 1. Flow diagram for included subjects.

Table 1

Patient characteristics (N = 252).

All patients Stratified by Antibiotics regimen P-value

		Narrow regimen	EGN regimen	FB regimen
Number of patients, N (%)	252 (100.0)	126 (50.0)	49 (19.4)	77 (30.6)
Median age (Q1-Q3), years Male, N (%)	34.0 (25.0-47.0) 216 (85.7)	33.0 (25.0-45.0) 108 (85.7)	38.0 (24.0-50.0) 40 (81.6)	32.0 (26.0-47.0) 68 (88.3)	0.5886 0.5796
Race, N (%) Caucasian	201 (79.8)	101 (80.2)	46 (93.9)	54 (70.1)	0.0052
Black	45 (17.9)	23 (18.3)	2 (4.1)	20 (26.0)
Hispanic	4 (1.6)	2 (1.6)	0 (0.0)	2 (2.6)
Other Location of injury, N (%) Lower extremity	2 (0.8) 177 (70.2)	0 (0.0) 97 (77.0)	1 (2.0) 28 (57.1)	1 (1.3) 52 (67.5)	0.0220
Upper extremity	59 (23.4)	26 (20.6)	16 (32.7)	17 (22.1)
Multiple	16 (6.3)	3 (2.4)	5 (10.2)	8 (10.4)
Median hemoglobin on arrival (Q1 – Q3) Missing data	13.2 (12.0-14.2) N = 5	13.1 (11.9-14.2) N = 3	13.3 (12.0-14.6) N = 2	13.1 (12.2-14.0) N = 0	0.7851
Median hematocrit on arrival (Q1-Q3) Median baseline pH (Q1-Q3)	38.9 (35.2-41.9) 7.36	39.0 (35.2-42.3) 7.36	38.1 (34.1-41.9) 7.34	39.0 (35.9-41.6) 7.37	0.4813 0.1771
	(7.31-7.40)	(7.30-7.40)	(7.18-7.37)	(7.33-7.40)
Missing data N = 163		N = 90	N = 34	N = 39
Median baseline white blood cell count (Q1-Q3) Missing data	12.4 (9.7-16.4) N = 6	12.4 (9.7-16.4) N = 3	13.5 (10.9-18.8) N = 3	11.6 (9.4-16.1) N = 0	0.2875
Injury severity score (ISS), median [Q1, Q3]	9.0 [4.0, 9.0]	9.0 [8.0, 9.0]	9.0 [4.0, 9.0]	9.0 [4.0, 9.2]	0.133
Median serum lactate (Q1-Q3), mmol/L Missing Data	1.8 (1.3-2.6) N = 248	NC N = 126	1.8 (1.3-2.6) N = 45	NC N = 77	NC
Median time to antimicrobial administration (Q1-Q3), hours	1.90 (0.53-7.78)	2.12 (0.63-8.90)	2.72 (0.65-6.25)	1.57 (0.30-8.02)	0.3642
(Q1-Q3) = 1st and 3rd quartiles; NC = not calculable. FB = Fluoroquinolone-based. EGN = Expanded gram-negative.

Fig. 2. Antimicrobial distribution.

Table 2

Patient outcomes (N = 252).

All patients Stratified by antibiotics regimen P-value

					Narrow regimen		EGN regimen		FB regimen
	Number of patients, N (%)		252 (100.0)		126 (50.0)		49 (19.4)		77 (30.6)
	Primary outcome
	Post-traumatic infection (any of the following), N (%)		3 (1.2)		1 (0.8)		2 (4.1)		0 (0.0)	0.1797
	Bacteremia		2 (0.8)		1 (0.8)		1 (2.0)		0 (0.0)	0.4442
	Cellulitis		0 (0.0)		0 (0.0)		0 (0.0)		0 (0.0)	1.0000
	Osteomyelitis		1 (0.4)		0 (0.0)		1 (2.0)		0 (0.0)	0.1944
	Secondary outcomes
	Median hospital length of stay (Q1 – Q3), days		3.0		3.0		5.0		2.0	b0.0001
			(2.0-6.0)		(2.0-5.0)		(3.0-9.0)		(1.0-4.0)
	90-day patient readmission, N (%)		225 (89.3)		112 (88.9)		45 (91.8)		68 (88.3)	0.8064
	MDR bacteria – any positive culture, N (%)		2 (0.8)		0 (0.0)		2 (4.1)		0 (0.0)	0.0372
	MRSA results, N (%)									0.0070
	None		249 (98.8)		126 (100.0)		46 (93.9)		77 (100.0)
	Positive for MRSA DNA		2 (0.8)		0 (0.0)		2 (4.1)		0 (0.0)
	Negative for MRSA DNA		1 (0.4)		0 (0.0)		1 (2.0)		0 (0.0)
	MRSA – positive or negative DNA results, N (%)		3 (1.2)		0 (0.0)		3 (6.1)		0 (0.0)	0.0070
	Number of drug-related ADRs documented									0.0533
	0		248 (98.4)		126 (100.0)		47 (95.9)		75 (97.4)
	1		0 (0.0)		0 (0.0)		0 (0.0)		0 (0.0)
	2		4 (1.6)		0 (0.0)		2 (4.1)		2 (2.6)
	Acute renal failure, N (%)		4 (1.6)		0 (0.0)		2 (4.1)		2 (2.6)	0.0533
	Hyperkalemia, N (%)		0 (0.0)		0 (0.0)		0 (0.0)		0 (0.0)	1.0000

(Q1 – Q3) = 1st and 3rd quartiles; NC = not calculable. FB = Fluoroquinolone-based.

EGN = Expanded gram negative.

MDR = Multi-drug resistant.

MRSA = Methicillin-resistant Staphylococcus aureus.

included a Staphylococcus epidermidis bacteremia, and a case of osteo- myelitis secondary to Actinomyces spp. One patient in the narrow regi- men suffered from a Staphylococcus epidermidis bacteremia, with an additional unidentified gram-positive bacterium also isolated.

Secondary outcomes

Patients receiving expanded gram-negative coverage in the form of an aminoglycoside or extended-spectrum cephalosporin, had a longer median hospital length of stay than patients in either narrow or FB reg- imens (5 days vs 3 days vs 2 days, p <= 0.0001). Additionally, more MDR bacteria (2 cases vs 0 cases vs 0 cases, p = 0.0372) and MRSA surveil- lance cultures (2 cases vs 0 cases vs 0 cases, p = 0.007) were positive in the EGN regimen than the other two cohorts. Two cases of acute renal failure were identified in the EGN and FB groups, however, this endpoint did not reach statistical significance. Similarly, two cases of documented Adverse drug effects were also seen in the EGN and FB reg- imens, but did not provide statistically significant differences when compared to the narrow regimen. Lastly, we were unable to identify any difference in 90-day patient readmissions between each of three co- horts and failed to isolate any instances of Clostridium difficile infection, hyperkalemia, or 30-day mortality.

Discussion

Civilian firearm violence is a significant healthcare burden in the United States, with estimated Medical costs surpassing $41 billion in 2010 [16]. Additive to the immediate mortality risk firearms pose, gunshot-related fractures are a major problem for clinicians due to their risk of underlying Vascular injury and infection. Exacerbating these issues is the lack of certainty surrounding the appropriate man- agement of these wounds. Although the decision for debridement and surgery are often dependent on the location and extent of the injury, an- timicrobials are almost uniformly provided. Unfortunately, although this practice is corroborated by both The Eastern Association for the Sur- gery of trauma guidelines, and the Surgical Infection Society Guideline, differing opinions on antimicrobial selection make interpretation diffi- cult for clinicians [5,6].

Given the heterogeneity in both guideline recommendations and clinical practice, we sought to evaluate the effect of a broad versus nar- row spectrum antimicrobial regimen for the prevention of post- traumatic infections in firearm open-fracture injuries. Our findings indi- cate that in patients with gunshot wound fractures isolated to an ex- tremity but requiring operative fixation, broad-spectrum gram- negative coverage with either an aminoglycoside, expanded coverage beta-lactam, or fluoroquinolone, was not associated with a reduction of post-traumatic infections when compared to a narrower spectrum antimicrobial agent. These findings align with another recent study by Lloyd et al. evaluating open-fracture antibiotic prophylaxis in combat- related open-fracture injuries [17]. In their investigation, combat open-fracture injuries treated with a narrow spectrum agent was asso- ciated with a similar risk of osteomyelitis development as an EGN regi- men. Although rates of superficial SSTI were higher in the narrow regimen, patients receiving EGN coverage had a greater number of bac- teria isolated that were resistant to aminoglycosides or fluoroquinolones.

We were unable to detect any differences in the rates of Adverse drug reactions, Clostridium difficile infections, 90-day readmission rates, 30-day mortality rates, hyperkalemia, or acute renal failure. Al- though the EGN regimens were not associated with any of these out- comes in our study, prior investigations have produced concern regarding the use of antimicrobial agents with expanded spectrums, and their association with the development of infection, resistance, and production of adverse drug events [17-20].

The fluoroquinolone class of antibiotics is of particular interest to cli- nicians due to their broad spectrum bacterial coverage, excellent oral

bioavailability, and long half-lives facilitating once or twice daily admin- istration. Although evidence is mixed regarding their benefit in open- fracture patient populations, concern has been raised regarding their in- hibitory effect on osteoclast and osteoblast activity, and negative associ- ation with fracture healing [21,22]. Furthermore, fluoroquinolone use has been identified as an independent risk factor for the development of MRSA colonization, and the FDA has recently strengthened boxed warnings for the drug class due to an increase in the numbers of tendinopathies, mental health side effects, hypoglycemic events, and aortic dissections that have been reported [23-28]. Although none of these events were documented in our study, the low infection rates in both the narrow-regimen and fluoroquinolone groups challenge any added utility of fluoroquinolone use in low-grade civilian gunshot frac- tures, especially given the risks that fluoroquinolone use poses.

Our analysis suggests that an EGN antimicrobial regimen was not as-

sociated with an improvement in the number of post-traumatic infec- tions. Although our reported infection rate of 1.2% was low, it closely mirrors that of other studies examining civilian gunshot open-fracture injuries [8,11,12,17]. Given these low infection rates, a larger sample size would have been needed to detect a significant difference between groups. Additional limitations include the lack of randomization and prescriptive assignment of antimicrobials, including treatment duration and time-to-antibiotic provision. Although not explicitly required, clini- cians at UKCMC follow the EAST guideline recommendations and com- monly utilize nursing and pharmacy staff to initiate antibiotics early in the patient’s stay. Due to the retrospective nature of our study it is pos- sible that patient comorbidities may have differed among patient co- horts, potentially confounding our primary and secondary outcomes. Lastly, it is possible that our initial UKTR query did not capture all gun- shot wounds presenting to our emergency department during the study period. Added strengths of our study include thorough inclusion and ex- clusion criteria generating relatively well-matched patient cohorts, in- cluding injury severity, and strong statistical methods.

Conclusion

In summary, for civilian patients experiencing a gunshot-related open-fracture, an expanded gram-negative antimicrobial regimen in the form of an aminoglycoside, extended-coverage beta-lactam, or fluo- roquinolone, were not associated with a reduction in post-traumatic in- fections. A prospective, randomized trial is needed to confirm these results.

Funding

The project described was supported by the NIH National Center for Advancing Translational Sciences through grant number UL1TR001998. The content is solely the responsibility of the authors and does not nec- essarily represent the official views of the NIH.

Declaration of Competing Interest

None to disclose.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2019.158358.

References

1. Fowler KA, et al. Firearm injuries in the United States. Prev Med 2015;79:5-14.
2. Centers for Disease Control, National Center for Injury Prevention and Control. WISQUARS fatal injuries: mortality reports. Available at http://webappa.cdc.gov/ sasweb/ncipc/mortrate.html, Accessed date: 14 November 2013.
3. Manring MM, et al. Treatment of war wounds: a historical review. Clin Orthop Relat Res 2009;467(8):2168-91.
4. Eardley WG, et al. Infection in conflict wounded. Philos Trans R Soc Lond B Biol Sci 2011;366(1562):204-18.
5. Hoff WS, et al. East practice management guidelines work group: update to practice management guidelines for Prophylactic antibiotic use in Open fractures. J Trauma 2011;70(3):751-4.
6. Hauser CJ, Adams Jr CA, Eachempati SR. Surgical infection society guideline: prophy- lactic antibiotic use in open fractures: an evidence-based guideline. Surg Infect (Larchmt) 2006;7(4):379-405.
7. Miclau T, Farjo LA. The antibiotic treatment of gunshot wounds. Injury 1997;28: C1-5.
8. Hospenthal DR, et al. Guidelines for the prevention of infections associated with combat-related injuries: 2011 update: endorsed by the Infectious Diseases Society of America and the Surgical Infection Society. J Trauma 2011;71(2 Suppl 2): S210-34.
9. Lichte P, et al. A civilian perspective on ballistic trauma and gunshot injuries. Scand J Trauma Resusc Emerg Med 2010;18:35.
10. C, L.O., M. M, and M.J. N. Controversies in the management of open fractures. Open Orthop J 2014;8:178-84.
11. Sathiyakumar V, et al. Gunshot-induced fractures of the extremities: a review of an- tibiotic and debridement practices. Curr Rev Musculoskelet Med 2015;8(3):276-89.
12. Volgas DA, Stannard JP, Alonso JE. Current orthopaedic treatment of ballistic injuries.

    Injury 2005;36(3):380-6.

    Murray CK, et al. Prevention of infections associated with combat-related extremity injuries. J Trauma 2011;71(2 Suppl 2):S235-57.

13. Tohira H, et al. Comparisons of the outcome prediction performance of injury sever- ity scoring tools using the abbreviated injury scale 90 update 98 (AIS 98) and 2005 update 2008 (AIS 2008). Ann Adv Automot Med 2011;55:255-65.
14. Harris Paul A, Taylor Robert, Thielke Robert, Payne Jonathon, Gonzalez Nathaniel, Conde Jose G. Research electronic data capture (REDCap) – a metadata-driven meth- odology and workflow process for providing translational research informatics sup- port. J Biomed Inform 2009;42(2):377-81 Apr.
15. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. Web-based injury statistics query and reporting system (WISQARS) [online]. Available from URL www.cdc.gov/injury/wisqars; 2015, Accessed date: 10 January 2019.

    Lloyd BA, et al. Early infectious outcomes after addition of fluoroquinolone or amino- glycoside to posttrauma antibiotic prophylaxis in combat-related open fracture inju- ries. J Trauma Acute Care Surg 2017;83(5):854-61.

16. Vasoo S, Barreto JN, Tosh PK. Emerging issues in Gram-negative bacterial resistance: an update for the practicing clinician. Mayo Clin Proc 2015;90(3):395-403.
17. Patolia S, et al. Risk factors and outcomes for multidrug-resistant gram-negative ba- cilli bacteremia. Ther Adv Infect Dis 2018;5(1):11-8.
18. Patzakis MJ, et al. Prospective, randomized, double-blind study comparing single- agent antibiotic therapy, ciprofloxacin, to combination antibiotic therapy in open fracture wounds. J Orthop Trauma 2000;14(8):529-33.
19. Huddleston PM, et al. Ciprofloxacin inhibition of experimental fracture healing. J Bone Joint Surg Am 2000;82(2):161-73.
20. Holtom PD, et al. Inhibitory effects of the quinolone antibiotics trovafloxacin, cipro- floxacin, and levofloxacin on osteoblastic cells in vitro. J Orthop Res 2000;18(5): 721-7.
21. Weber SG, et al. Fluoroquinolones and the risk for methicillin-resistant Staphylococ- cus aureus in hospitalized patients. Emerg Infect Dis 2003;9(11):1415-22.
22. Salangsang JA, et al. Patient-associated risk factors for acquisition of methicillin- resistant Staphylococcus aureus in a tertiary care hospital. Infect Control Hosp Epidemiol 2010;31(11):1139-47.
23. Dziekan G, et al. methicillin-resistant Staphylococcus aureus in a teaching hospital: investigation of nosocomial transmission using a matched case-control study. J Hosp Infect 2000;46(4):263-70.
24. Levaquin (levofloxacin) [package insert]. Austin, Tx: Hospira, Inc; 1996.
25. FDA updates warnings for fluoroquinolone antibiotics on risks of mental health and low Blood sugar adverse reactions. Available at https://www.fda.gov/newsevents/ newsroom/pressannouncements/ucm612995.htm; 2016, Accessed date: 10 January

    2019.

    FDA warns about increased risk of ruptures or tears in the aorta blood vessel with fluoroquinolone antibiotics in certain patients. Available at https://www.fda.gov/ Drugs/DrugSafety/ucm628753.htm; 2018, Accessed date: 10 January 2019.