Liquid chromatography–tandem mass spectrometry for rapid and selective simultaneous determination of fluoroquinolones level in human aqueous humor

Accepted Manuscript Liquid chromatography–tandem mass spectrometry for rapid and selective simultaneous determination of fluoroquinolones level in human aqueous humor

Ala A. Alhusban, Ola A. Tarawneh, Sanabel O. Dawabsheh, Ahmad A. Alhusban, Feras W. Abumhareb PII: DOI: Reference:

S1056-8719(18)30725-1 https://doi.org/10.1016/j.vascn.2019.03.001 JPM 6570

To appear in:

Journal of Pharmacological and Toxicological Methods

Received date: Revised date: Accepted date:

28 August 2018 23 December 2018 14 March 2019

Please cite this article as: A.A. Alhusban, O.A. Tarawneh, S.O. Dawabsheh, et al., Liquid chromatography–tandem mass spectrometry for rapid and selective simultaneous determination of fluoroquinolones level in human aqueous humor, Journal of Pharmacological and Toxicological Methods, https://doi.org/10.1016/j.vascn.2019.03.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Liquid chromatography–tandem mass spectrometry for rapid and selective simultaneous determination of fluoroquinolones level in human aqueous humor Ala A. Alhusbana,* [email protected], Ola A. Tarawneha, Sanabel O. Dawabsheha, Ahmad A. Alhusbanb, Feras W. Abumharebb Department of Pharmacy, Faculty of Pharmacy, Al-Zaytoonah University of Jordan, Amman, Jordan bDepartment of Ophthalmology, Royal Medical Services; Ophthalmologist, King Talal Military Hospital, Almafraq, Jordan *

SC RI P

T

a

Corresponding author.

Abstract

AC

CE

PT

ED

MA

NU

Endophthalmitis, an intraocular infection, may lead to irreversible loss of vision. Antimicrobial chemotherapy is prescribed prior to ocular surgical procedures to avoid endophthalmitis. Fluoroquinolones are the most commonly prescribed and used antibiotics during such procedures. However, the selection of a single entity and proper regimen is not specified in medical guidelines. The objective of this study is to develop a rapid and selective simultaneous high-performance liquid chromatography-tandem mass spectroscopic method to explore the bioavailability of 4 fluoroquinolones, including 0.5% moxifloxacin hydrochloride, 0.3% gatifloxacin, 0.3% ciprofloxacin hydrochloride, and 0.3% ofloxacin, in human aqueous humor after antibiotic topical administration using gemifloxacin as Internal Standard according to the European Medicines Agency (EMA) and US Food and Drug Administration (FDA) guidelines. The newly validated method was capable of accurately and precisely quantifying each antibiotic at the lowest reported lower limit of quantification of 10 ng/mL and operating on a very low pipetting volume of 15 μL, indicating a reliable quantitation of all analytes simultaneously using a very small quantity of aqueous humor with total chromatographic run time of 2.5 minutes. Sixty-seven patients were divided into 4 groups for each antibiotic. Before the cataract surgery, each group received 4 drops of one of the antibiotics over 1 hour, separated by 15 minutes time interval. After 15 - 20 minutes from the last drop, approximately 50–100 μL of aqueous humor was collected during surgery for analysis. The average concentrations of moxifloxacin, gatifloxacin, ofloxacin and ciprofloxacin in aqueous humor samples were 891.8, 271.7, 191.4 and 69.5 ng/mL, respectively. Only moxifloxacin average concentration was higher than the minimum inhibitory concentration for the common endophthalmitis pathogens.

Keywords: fluoroquinolones, endophthalmitis, aqueous humor, HPLC–MS/MS, Ophthalmic solution, cataract

ACCEPTED MANUSCRIPT 1. Introduction

SC RI P

T

Till date, there has been no standardized procedure to minimize the incidence of endophthalmitis before a cataract surgery in the USA and Canada (Olson, et al., 2017). Practices that lower the risk of this devastating complication that may lead to vision loss show huge disparities between countries due to the limited records from randomized clinical trials (Villegas, Schwartz, Grzybowski, Relhan, & Flynn, 2018). For example, the surgeon may use sub-conjunctival injections, intracameral or infusion fluid antibiotics before the surgery (Gower, Lindsley, Nanji, Leyngold, & McDonnell, 2013). However, to be effective, the antibiotic should be available at the site of action in a concentration higher than the Minimum Inhibitory Concentration (MIC) for the pathogens responsible for endophthalmitis (Mah, 2004). Moreover, some surgeons may use only povidoneiodine wash in addition to outer surface sterilization of the eye, while others may use topical antibiotic followed by povidone-iodine wash (Grzybowski, Kuklo, Pieczynski, & Beiko, 2016), which has shown a synergic bacterial inhibition effect (de Kaspar, et al., 2008).

ED

MA

NU

Chang et al. analyzed the data of a survey completed in 2014 by the American Society of Cataract and Refractive Surgery (ASCRS) members concerning post-operative endophthalmitis prophylaxis manoeuvres and compared it with the same survey conducted in 2007. Majority of health providers were seen to prefer pre-operative topical fluoroquinolones application. Besides, a tendency towards the usage of ofloxacin and ciprofloxacin over fourth-generation gatifloxacin and moxifloxacin was noticed since 2014. This tendency was justified due to the low cost of ofloxacin and ciprofloxacin, generic availability, comparable efficacy, and resistance considerations (Chang, Braga-Mele, Henderson, Mamalis, & Vasavada, 2015). Nevertheless, a higher intraocular penetration of topical fluoroquinolone is required for the drug to be effective (Chung, et al., 2013).

AC

CE

PT

Fluoroquinolones were introduced in the market in 1960, they hold improved safety profile and better in vivo efficacy against resistant bacteria compared to other antimicrobial agents (Dudhatra, et al., 2013). Furthermore, fluoroquinolones have been considered by ophthalmologists as the standard topical antibiotic for ophthalmologic infection prevention and treatment due to their broad spectrum, rapid onset of action, high tolerability and low toxicity profile (Mah, 2004). However, the choice of a single entity of the fluoroquinolones family and administration regimen is not a wellestablished one. However, it depends on the ophthalmologist’s best practices, available antibiotics, and economical consideration (Ranganath & Bansal, 2017). Hence, intraocular penetration for each of the commonly used fluoroquinolone antibiotics should be evaluated to ensure the effectiveness of the followed regimen in endophthalmitis prevention. Chromatographic techniques were the only techniques employed for the measurement of fluoroquinolones aqueous humor penetration. Reported methods include highperformance liquid chromatography with fluorescence detection (Beck, Fischer, Hehl, Guthoff, & Drewelow, 1997; Fukuda, et al., 2012; McCulley, Caudle, Aronowicz, & Shine, 2006; Yamada, Ishikawa, Mochizuki, & Kawai, 2006) and HPLC/MS (Bucci, 2004; Bucci Jr, Nguimfack, & Fluet, 2016; Domingos, et al., 2017). The liquid chromatographic system hyphenated to tandem mass spectroscopy (HPLC-MS/MS) facilitates the technology to detect similar or different molecules from the same or different family in

ACCEPTED MANUSCRIPT

T

a simultaneous and selective manner. This helps to reduce the resources, time, and number of methods required for the quantitation of different molecules (Remane, Meyer, Wissenbach, & Maurer, 2010). HPLC-MS/MS has only been used in two studies. In the first study, it was utilized to investigate the absorption of moxifloxacin into human aqueous humor after the administration of moxifloxacin hydrochloride ophthalmic solution (Katz, et al., 2005). In the second study, it was applied to investigate the bioavailability of moxifloxacin and gatifloxacin in aqueous humor during cataract surgery (Kim, Stark, & O'Brien, 2005). To the best of our knowledge, HPLCMS/MS has not been reported for simultaneous determination of fluoroquinolones in human ocular samples.

SC RI P

The objective of this study was to develop and validate a novel bioanalytical method for simultaneous determination of moxifloxacin, gatifloxacin, ciprofloxacin, and ofloxacin using HPLC-MS/MS system, which can easily be applied to other fluoroquinolones. The additional objective was to compare the bioavailability concentration of the antibiotics with the MIC for endophthalmitis pathogens to determine the efficacy of the followed regimens. Which is valuable to the patients in terms of effective post-operative endophthalmitis prevention.

NU

2. Materials and methods 2.1. Reagents and chemicals

CE

PT

ED

MA

Gatifloxacin (purity > 98%) was purchased from ACROS (New Jersey, USA), ofloxacin (purity > 99%) was purchased from Sigma-Aldrich (St. Louis, MO, USA), moxifloxacin hydrochloride (purity > 98%) was obtained from SPIMACO ADDWAEIH (Kuwait, Kuwait), and ciprofloxacin hydrochloride (purity > 98%) was obtained from United Pharmaceuticals (Amman, Jordan). HPLC grade methanol was purchased from Merck (Darmstadt, Germany), ultra-pure deionized water from MilliQ® water purification instrument (Millipore, Bedford, MA, USA), ammonium hydroxide from Scharlau (Barcelona, Spain), ammonium formate from Sigma-Aldrich (St. Louis, MO, USA), and formic acid from TEDIA (Fairfield, OH, USA). Blank aqueous humor matrix used in method development and validation was obtained from patients who underwent cataract surgery, where povidone-iodine was only used as the prophylactic procedure.

AC

2.2. Instrumentation and HPLC–MS/MS analytical conditions Agilent HPLC (USA) was interfaced with AB Sciex QTRAP 4500 mass spectroscopy (AB Sciex Instruments, Foster City, CA, USA). The analytical column used was Eclipse XDBC18 (4.6 X 150 mm, 5 µm). The column was eluted at 40 °C with Methanol: Ammonium Formate Buffer (60:40) and 1 mL of formic acid/L in isocratic conditions. Validated Analyst® Software version 1.6.3., AB Sciex (USA) was used for system control, samples acquisitions and data analysis. The ion transitions were monitored on m/z 332.200 →314.200, m/z 376.200→332.300, m/z 390.000→372.300, m/z 402.100→384.200 and m/z 362.100→318.100 for ciprofloxacin, gatifloxacin, Gemifloxacin (IS), moxifloxacin and ofloxacin respectively. Table 1 indicates the compound and source –dependent parameters specific to the developed method. A positive electrospray interface was used to introduce the mobile phase carrying the analytes and IS to the mass spectroscopy. Multiple reaction monitoring was used with signal transition (Precursor ion (Q1)  Daughter ion (Q3)).

ACCEPTED MANUSCRIPT 2.3. Calibration standards and quality control (QC) samples preparation Standard stock solutions of 500 µg/mL of ciprofloxacin and ofloxacin were prepared in methanol: ammonium hydroxide (1.5:25), gatifloxacin and gemifloxacin (IS) in methanol: deionized water (50:50) whereas moxifloxacin in methanol only. All prepared solutions were stored at 4 °C.

SC RI P

T

Analytes’ working solutions were prepared daily by diluting each of the aforementioned analytes’ stock solutions and completing the volume with methanol: water (50:50). These analytes’ working solutions were used to first establish calibration curve points in drug-free aqueous humor samples to obtain standard concentrations of 10, 20, 50, 100, 200, 400, 700, and 1000 ng/mL of each analyte and to prepare QC samples at 3 levels as low QC (30 ng/mL), mid QC (500 ng/mL), and high QC (850 ng/mL). 2.4. Aqueous humor samples collection and preparation

MA

NU

Ophthalmologists collected approximately 50–100 μL of aqueous humor by paracentesis using a 30-gauge needle on a tuberculin syringe during the cataract surgery. Samples were kept in pre-labeled syringes and stored at nominal temperature of –70 °C until analysis. Aqueous humor samples just before analysis were prepared by spiking 15 µL of the aqueous humor sample with 15 µL of the IS solution and then diluting the sample in 500 µl with the mobile phase. 2.5. Method validation

PT

2.5. 1. Selectivity

ED

The analytical method was validated according to the FDA and EMA guidelines for bioanalytical method validation.

AC

CE

Selectivity with respect to methods is important to differentiate between the target analytes from endogenous substances and other compounds in human aqueous humor samples. Selectivity of the developed method was evaluated using 6 sources of aqueous humor by comparing the peak signals at the target analytes retention times in blank extracted samples with the peak signals at the target analytes retention times in Lower Limit of Quantitation (LLOQ) samples. The peak signal of the blank matrix at the target analyte retention times should be less than 20% of the peak signal of the LLOQ sample, while it should be less than 5% at the retention time of the IS. 2.5. 2. Calibration curve and linearity Three calibration curves covering the range 10–1000 ng/mL for all analytes were prepared to validate the method linearity. For each calibration curve, eight non-zero calibrators were prepared of 10, 20, 50, 100, 200, 400, 700, and 1000 ng/mL of each analyte in addition to blank and zero samples. The back calculated concentration should be within 15% for all the calibrators, except for the LLOQ, where it should be less than or equal to 20%. Seventy-five percent (75%) of the calibration curve points should pass the acceptance criteria, including the LLOQ and ULOQ with a correlation coefficient of 0.99 or greater.

ACCEPTED MANUSCRIPT 2.5. 3. Precision and accuracy Within-run accuracy and precision were evaluated at 6 replicates of 4 QC levels, 10, 30, 500 and 850 ng/mL of each analyte in 1 analytical run. Between-run accuracy and precision were evaluated from 6 replicates of each QC level from 3 different days. In order to accept the results, the mean accuracy and precision was required to be within 15% of the nominal concentration except for the LLOQ, which should be within 20% of the nominal concentration.

T

2.5. 4. Matrix effect

SC RI P

Matrix effect was validated to investigate co-eluting compounds suppression or enhancement effect on the signal peak. Matrix effect was evaluated over two levels— low QC and high QC—by comparing the instrument response of 6 samples of pure solution of analytes to instrument the responses of 6 extracted aqueous humor batches. For the successful implementation of the matrix effect, IS normalized matrix factor CV% should not be more than 15 %.

NU

2.5. 5. Stability

MA

The stability of samples at room temperature was evaluated by preparing 6 samples in sufficient amounts at low and high QC levels for each analyte and analyzing them immediately against a calibration curve. The remaining samples were stored at room temperature for 6 hours. Then, samples were prepared and analyzed.

PT

ED

Freeze-thaw stability was evaluated by preparing sufficient amounts of the 6 samples at low and high QC levels for each analyte and analyzing them immediately against a calibration curve; the remaining volume was stored at a nominal temperature of -70 °C for 24 hours. After that, these samples were left at room temperature to thaw, prepare, and analyze.

CE

The accuracy of the samples was calculated by comparing the actual concentration with the nominal concentration. The calculated concentration should be within 15% of the nominal concentration. 2.5. 6. Lower limit of quantification (LLOQ)

AC

The LLOQ is the lowest reliable concentration in the calibration curve that could be quantified by the analytical method. The LLOQ for each analyte was 10 ng/mL. In order to validate the LLOQ of the method, the analyte signal at the analyte retention time of an extracted blank matrix was compared to the analyte signal at the same retention time of an LLOQ sample prepared from the same matrix. The analyte signal at the blank sample should be less than 20% of the analyte signal at the LLOQ sample. 2.6. Pharmacokinetic study Sixty-seven patients were recruited in a randomized, open-label, one-site bioavailability study of 0.3% ciprofloxacin hydrochloride, 0.3% gatifloxacin, 0.5% moxifloxacin hydrochloride and 0.3% ofloxacin (w/v) ophthalmic solutions in patients’ aqueous humor samples undergoing cataract surgery. Ethics approval (approval number 8/2017) was obtained from the Human Research Ethics Committee in the Royal Medical

ACCEPTED MANUSCRIPT

3. Results and discussion 3.1. Method development and validation

SC RI P

T

Services, Jordan Armed Forces, before the study. The patients were asked for their consent before the commencement of the study, and their written approval was obtained. Patients were divided into 4 groups for the administration of each antibiotic. Before the surgery, each group received 4 drops of one of the antibiotics over 1 hour separated by 15 minutes time interval. Patients who missed any of the 4 doses were excluded from the study. Just after 15 - 20 minutes from the last dose, around 50–100 μL of aqueous humor was collected during the surgery from each patient for analysis. The validated method was applied to simultaneously quantify the concentration of ciprofloxacin, gatifloxacin, moxifloxacin, and ofloxacin in the patients’ human aqueous humor samples. Analytical method development and validation and biological samples analysis took place in the Arab Company for Drug Industries & Medical Appliances Center for Bioequivalence and Clinical Studies (ACDIMA BioCenter), Amman, Jordan.

CE

PT

ED

MA

NU

The development and validation of an analytical method for simultaneous determination of all target analytes have met the acceptance criteria for the applicable FDA ("Food and Drug Administration, Guidance for industry: bioanalytical method validation," 2001) and EMA ("European Medicines Agency, Guidance on Bioanalytical Method Validation," 2011 ). Moreover, no carry over was reported for any of the analytes or the IS. Also, the sample processing involved only a simple and effective dilution procedure. Further, simultaneous chromatographic peaks separation was sensitive, selective, accurate, and reproducible across all analytes, despite the very challenging LLOQ of 10 ng/mL and low pipetting volume of 15 µL within a run time of 2.5 minutes. The 10 ng/mL LLOQ and 15 µl sample volume are the lowest LLOQ and sample volume employed for aqueous humor respectively to the best of our knowledge. The lowest sample volume reported in the literature was 50 µl of human aqueous humor and was used for the determination of moxifloxacin (Bucci Jr, et al., 2016; Katz, et al., 2005). A recent research reported the usage of aliquot of 500 μL of aqueous humor for the simultaneous determination of gatifloxacin, moxifloxacin, and besifloxacin (Domingos, et al., 2017). Likewise, the LLOQ was reported at 40 ng/mL for ciprofloxacin (Bucci, 2004), 190 ng/ml for moxifloxacin (Bucci Jr, et al., 2016), and 420 ng/mL for gatifloxacin and moxifloxacin (Domingos, et al., 2017).

AC

3.1. 1. Selectivity

The method was selective, since no interfering peaks at the retention time of any of the analytes or IS were observed. This has been illustrated in Figure 1, which shows the instrument response to 5 different blank human aqueous humor samples (a–e) and to LLOQ of analytes spiked to human aqueous humor samples (f–j). 3.1. 2. Calibration curve and linearity The method showed good linearity over the analytical range of 10.0-1000.0 ng/mL with an LLOQ of 10 ng/mL for all analytes in aqueous humor samples . Calibration curves met the acceptance criteria as per the FDA and EMA guidelines. Mean calibration curve parameters and accuracy of calibrators are listed in Table 2 and

ACCEPTED MANUSCRIPT Table 3 respectively. The correlation factor was equal or above 0.999 for all analytes, and the accuracy range was between 93.86–109.11 % for all analytes. 3.1. 3. Precision and accuracy

T

Within- and between-run accuracy and precision were also within the acceptance criteria; results obtained are summarized in Table 4 and Table 5, repectively. The %CV of within- and between-run precision were less than 15% and the average accuracy values were within the range of 80%–120%. This finding confirms that the newly developed method is accurate, reliable and reproducible.

SC RI P

3.1. 4. Matrix effect

There was no significant suppression or enhancement of signals detection of the analyte and IS based on the IS-Normalized matrix factor results. IS-Normalized matrix factor CV% values were less than 15% that indicate reproducible and consistent results. The results are listed in Table 6.

NU

3.1. 5. Stability

AC

CE

PT

ED

MA

A short-term stability study was conducted for 6 hours at room temperature and samples were found to be stable. The calculated concentrations were within 15% of the nominal concentration. Data on short-term stability after 6 hours are listed in

MA

NU

SC RI P

T

ACCEPTED MANUSCRIPT

Table 7.

PT

ED

Freeze-thaw stability for 1 cycle was evaluated and found to be stable. The accuracy of the samples was within 15% of the nominal concentration. Data on freeze-thaw stability are listed in Table 8. 3.2. Fluoroquinolones aqueous humor concentrations

AC

CE

Cataract is reported to be one of the main causes of blindness in the Jordanian population (Hamarneh, 2013). A published study in 2015 has shown that untreated cataract is considered to be the major cause of blindness in northern Jordan, accounting for 46.7% of total cases of blindness (Rabiu, Al Bdour, Abu Ameerh, & Jadoon, 2015). However, specific medical guidelines are neither followed in the Royal Medical Services nor globally. Thus, on our personal investigation in the design of this project, the consultants affirmed their commitment to follow a specified procedure in order to minimize the incidence of endophthalmitis. The prophylaxis procedure involves the administration 4 drops of one of the fluoroquinolone antibiotics, moxifloxacin, gatifloxacin, ciprofloxacin, and ofloxacin 1 hour before the surgery, separated by a 15minutes interval, followed by washing the eye sac with povidone-iodine immediately before the surgery. This research was carried out on cataract patients, who decided to have surgery. Sixtyseven patients consented to the surgeon, who explained the study aims and protocol. All patients approved to be a part of the study and gave their written approval by signing the Informed Consent Form. Subsequently, they were divided into four groups: 27 patients received moxifloxacin eye drops, 16 patients received gatifloxacin eye drops,

ACCEPTED MANUSCRIPT

T

13 patients received ciprofloxacin eye drops and 11 patients received ofloxacin eye drops. A summary of the patients’ aqueous humor concentrations in ng/mL after topical administration of ciprofloxacin, gatifloxacin, moxifloxacin, and ofloxacin ophthalmic drops, including mean, minimum, and maximum concentration is listed in Table 9. As can be observed, there was high range of variation between minimum and maximum concentrations of all 4 fluoroquinolones in patients’ aqueous humor. This was expected due to the diversity of patients undergoing the cataract surgery (age, gender, race, disease conditions and associated chronic diseases). All samples were collected at fixed range of time (15 – 20) minutes after the last dose of the antibiotic used. The highest level in aqueous humor was for moxifloxacin, followed by gatifloxacin, ofloxacin, and finally ciprofloxacin (Figure 2).

AC

CE

PT

ED

MA

NU

SC RI P

Table 10 summarizes the MIC values of ciprofloxacin, gatifloxacin, moxifloxacin, and ofloxacin for the main pathogens, responsible mainly for endophthalmitis. Additionally, the comparison of the 4 fluoroquinolones mean concentrations in patients’ aqueous humor to the MIC levels of the most common pathogens causing endophthalmitis are shown in Figure 3. The average concentration of ciprofloxacin was 70 ng/mL in the patients’ aqueous humor samples. This means that the MICs were not achieved for any investigated pathogens. However, two individual values were above the MIC for Bacillus species only. A similar observation was made with regard to ofloxacin and ciprofloxacin. The average concentration of ofloxacin was observed to be 190 ng/mL in patients’ aqueous humor samples, which is higher than ciprofloxacin but lower than the MICs for the tested pathogens. However, the concentration of one patient’s aqueous humor sample exceeded the MIC for only Bacillus species. Gatifloxacin average concentration was 270 ng/mL in patients’ aqueous humor samples. This was higher than the MICs for Susceptible Staphylococcus aureus, Streptococcus pneumonia and Bacilli species, but not for Enterococci species. Nevertheless, the individual aqueous humor concentration of gatifloxacin was higher than the MICs of Enterococci species in the samples of 4 patients out of 16. The average concentration of moxifloxacin was 890 ng/mL in patients’ aqueous humor samples. This concentration was higher than the MICs for all the pathogens. Thus, the detected level of moxifloxacin was higher than other fluoroquinolones antibiotics. This may be attributed to the fact that moxifloxacin is more lipophilic in nature and subsequently penetrates the cornea to the eye tissue and accumulates there in higher concentration compared to the investigated fluoroquinolones (McCulley, et al., 2006). The increase in the concentration in the aqueous humor would enable the antibiotic to reach a concentration beyond the MIC in order to inhibit infections. It was found that the average concentration of the investigated second-generation fluoroquinolones in the patients’ samples was lower than the reported MICs for pathogens using the assigned procedure. This high MIC of ciprofloxacin and ofloxacin is attributed to the development of resistant pugs against second generations. Only one genetic mutation can provide bacterial resistance for second-generation fluoroquinolone (Mather, Karenchak, Romanowski, & Kowalski, 2002). Fourthgeneration fluoroquinolones, gatifloxacin, and moxifloxacin showed better inhibition activities than the second-generation fluoroquinolones, since they were designed to be resistant to expected mutation (Mather, et al., 2002). In the light of preceding data and our findings, moxifloxacin exhibited be tter activity in this procedure than other investigated fluoroquinolones. Furthermore, the employed

ACCEPTED MANUSCRIPT regimen in this study was well tolerated by the patients, and no adverse events were reported during or after the surgery. 4. Conclusion

MA

NU

SC RI P

T

Endophthalmitis is a cataract surgery complication. Despite the possibility that it may cause irreversible vision loss, there is no medical guideline for pre-operative prophylaxis measures. The usage of the prophylactic measures depends on the surgeon’s best practices and antibiotic’s availability. In order to be effective in endophthalmitis prevention, the concentration of the antibiotic in the aqueous humor should be higher than the pathogens’ MIC. In this study, 67 patients were recruited and subsequently divided into 4 groups; each group received one of the antibiotics, including ciprofloxacin, gatifloxacin, moxifloxacin, and ofloxacin. During the surgery, the consultants collected 50–100 µL of the patient’s aqueous humor and stored until analysis. A novel, rapid, selective, accurate, precise, and simultaneous method was developed and validated, applying HPLC-MS/MS system for determination of antibiotic concentration in patients’ aqueous humor samples. The results showed that moxifloxacin has higher bioavailability, followed by gatifloxacin, ofloxacin, and ciprofloxacin. Since moxifloxacin is the only antibiotic that has a higher concentration than the MIC for the most common pathogens, it is suggested that moxifloxacin is the antibiotic of choice for preventing post-operative endophthalmitis, following the aforementioned regimen. Acknowledgments

PT

ED

This work was supported by the Deanship of Scientific Research and Graduate Studies at Al-Zaytoonah University of Jordan (2015/10/34) and (2018-2017/28/19). The authors acknowledge the Ophthalmology Department at Royal Medical Services, Jordanian Armed Forces for assistance in the clinical part of the study. And ACDIMA BioCenter, Amman, Jordan for their help in the analytical part.

AC

References

CE

Competing interests The authors declare no competing interests.

Beck, R., Fischer, U., Hehl, E., Guthoff, R., & Drewelow, B. (1997). Penetration of ciprofloxacin, norfloxacin and ofloxacin into the aqueous humours of patients by different topical application modes. European journal of clinical pharmacology, 53, 251-255. Bucci, F. A. (2004). An in vivo study comparing the ocular absorption of levofloxacin and ciprofloxacin prior to phacoemulsification. American journal of ophthalmology, 137, 308-312. Bucci Jr, F. A., Nguimfack, I. T., & Fluet, A. T. (2016). Pharmacokinetics and aqueous humor penetration of levofloxacin 1.5% and moxifloxacin 0.5% in patients undergoing cataract surgery. Clinical ophthalmology (Auckland, NZ), 10, 783. Chang, D. F., Braga-Mele, R., Henderson, B. A., Mamalis, N., & Vasavada, A. (2015). Antibiotic prophylaxis of postoperative endophthalmitis after cataract surgery:

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI P

T

results of the 2014 ASCRS member survey. Journal of Cataract & Refractive Surgery, 41, 1300-1305. Chung, J. L., Lim, E. H., Song, S. W., Kim, B. Y., Lee, J. H., Mah, F. S., & Seo, K. Y. (2013). Comparative intraocular penetration of 4 fluoroquinolones after topical instillation. Cornea, 32, 1046-1051. de Kaspar, H. M., Kreutzer, T. C., Aguirre-Romo, I., Ta, C. N., Dudichum, J., Bayrhof, M., Klauss, V., & Kampik, A. (2008). A prospective randomized study to determine the efficacy of preoperative topical levofloxacin in reducing conjunctival bacterial flora. American journal of ophthalmology, 145, 136-142. e132. Domingos, L. C., Moreira, M. V. L., Keller, K. M., Viana, F. A. B., Melo, M. M., & Soto -Blanco, B. (2017). Simultaneous quantification of gatifloxacin, moxifloxacin, and besifloxacin concentrations in cornea and aqueous humor by LC-QTOF/MS after topical ocular dosing. Journal of pharmacological and toxicological methods, 83, 87-93. Dudhatra, G., Kumar, A., Awale, M., Kamani, D., Chauhan, B., Patel, H., & Mody, S. (2013). Determination of orbifloxacin in sheep plasma by high performance liquid chromatography with ultraviolet detection after intravenous and intramuscular administration. Journal of pharmacological and toxicological methods, 67, 50-53. European Medicines Agency, Guidance on Bioanalytical Method Validation. (2011 ). accessed 10/7/18, http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline /2011/2008/WC500109686.pdf. Food and Drug Administration, Guidance for industry: bioanalytical method validation. (2001). accessed 11/7/18, https://www.fda.gov/downloads/drugs/guidances/ucm070107.Pdf. Fukuda, M., Yamada, M., Kinoshita, S., Inatomi, T., Ohashi, Y., Uno, T., Shimazaki, J., Satake, Y., Maeda, N., & Hori, Y. (2012). Comparison of corneal and aqueous humor penetration of moxifloxacin, gatifloxacin and levofloxacin during keratoplasty. Advances in therapy, 29, 339-349. Gower, E. W., Lindsley, K., Nanji, A. A., Leyngold, I., & McDonnell, P. J. (2013). Perioperative antibiotics for prevention of acute endophthalmitis after cataract surgery. Cochrane Database Syst Rev, 7. Grzybowski, A., Kuklo, P., Pieczynski, J., & Beiko, G. (2016). A review of preoperative manoeuvres for prophylaxis of endophthalmitis in intraocular surgery: topical application of antibiotics, disinfectants, or both? Current Opinion in Ophthalmology, 27, 9-23. Hamarneh, S. (2013). Causes of blindness in the Jordanian population. Middle East Journal of Internal Medicine, 6. Katz, H. R., Masket, S., Lane, S. S., Sall, K., Orr, S. C., Faulkner, R. D., McCue, B. A., & Dahlin, D. C. (2005). Absorption of topical moxifloxacin ophthalmic solution into human aqueous humor. Cornea, 24, 955-958. Kim, D. H., Stark, W. J., & O'Brien, T. P. (2005). Ocular penetration of moxifloxacin 0.5% and gatifloxacin 0.3% ophthalmic solutions into the aqueous humor following topical administration prior to routine cataract surgery. Current medical research and opinion, 21, 93-94. Mah, F. S. (2004). Fourth-generation fluoroquinolones: new topical agents in the war on ocular bacterial infections. Current Opinion in Ophthalmology, 15, 316-320.

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

MA

NU

SC RI P

T

Mather, R., Karenchak, L. M., Romanowski, E. G., & Kowalski, R. P. (2002). Fourth generation fluoroquinolones: new weapons in the arsenal of ophthalmic antibiotics1. American journal of ophthalmology, 133, 463-466. McCulley, J., Caudle, D., Aronowicz, J., & Shine, W. (2006). Fourth-generation fluoroquinolone penetration into the aqueous humor in humans. Ophthalmology, 113, 955-959. Olson, R. J., Braga-Mele, R., Chen, S. H., Miller, K. M., Pineda, R., Tweeten, J. P., & Musch, D. C. (2017). Cataract in the Adult Eye Preferred Practice Pattern®. Ophthalmology, 124, P1-P119. Rabiu, M. M., Al Bdour, M. D., Abu Ameerh, M. A., & Jadoon, M. Z. (2015). Prevalence of blindness and diabetic retinopathy in northern Jordan. European journal of ophthalmology, 25, 320-327. Ranganath, A., & Bansal, A. (2017). Prophylaxis of postoperative endophthalmitis after cataract surgery. Expert Review of Ophthalmology, 12, 261-268. Remane, D., Meyer, M. R., Wissenbach, D. K., & Maurer, H. H. (2010). Ion suppression and enhancement effects of co‐eluting analytes in multi‐analyte approaches: systematic investigation using ultra‐high‐performance liquid chromatography/mass spectrometry with atmospheric‐pressure chemical ionization or electrospray ionization. Rapid Communications in Mass Spectrometry, 24, 3103-3108. Villegas, V. M., Schwartz, S. G., Grzybowski, A., Relhan, N., & Flynn, H. W. (2018). Endophthalmitis Prophylaxis: Different Practices from Around the World. In Endophthalmitis (pp. 345-356): Springer. Yamada, M., Ishikawa, K., Mochizuki, H., & Kawai, M. (2006). Corneal penetration of simultaneously applied topical levofloxacin, norfloxacin and lomefloxacin in human eyes. Acta Ophthalmologica, 84, 192-196.

ACCEPTED MANUSCRIPT Table 1: Compound and source –dependent parameters Manually tuned parameters value for Ciprofloxacin, Gatifloxacin, Moxifloxacin and Ofloxacin Q1 (m/z)

Q3 (m/z)

DP (Volt)

CE (Volt)

Ciprofloxacin

332.2

314.2

106.00

26.00

Gatifloxacin

376.2

332.3

113.02

25.00

Moxifloxacin

402.1

384.2

104.00

30.00

Ofloxacin

362.1

318.1

96.0

SC RI P

T

Analyte

26.00

Other Parameters:

21.00

CAD

Medium

TEM

600.00

GS1

35.00

GS2

35.00

EP

10

CXP

15.00

NU

CUR

AC

CE

PT

ED

MA

CAD: CAD Gas, CE: Collision Energy, CUR: CUR, CXP: Collision Cell Exit Potential, DP: Declustering Potential, EP: Entrance Compound Potential, GS2: Ion Source Gas 1, GS2: Ion source Gas 2, Q1: precursor ion, Q3: daughter ion, TEM: Temperature.

Table 2: Mean calibration curve parameters of ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin calibration curves

ACCEPTED MANUSCRIPT Parameter

Ciprofloxacin Gatifloxacin Moxifloxacin

Ofloxacin

Slope

0.000452

0.000421

0.000693

0.001583

B

Intercept

0.000501

0.000153

-0.000009

0.000727

R

Correlation factor

0.9990

0.9992

0.9990

0.9990

SC RI P

T

A

Table 3: Mean accuracy of ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin calibrators (%). The concentration in (10 ng/mL).

10.0

94.71

93.86

97.14

94.27

2

20.0

101.18

101.61

100.60

100.50

3

50.0

107.63

107.31

109.11

108.33

4

100.0

99.10

99.06

97.18

97.46

5

200.0

97.12

96.65

93.86

98.53

6

400.0

102.14

103.14

103.13

103.05

7

700.0

95.54

97.34

97.18

96.35

8

1000.0

102.58

101.04

101.79

101.51

CE

PT

ED

1

MA

NU

Sample Ciprofloxacin Gatifloxacin Moxifloxacin Ofloxacin Nominal % % % % conc.

AC

Sample No.

Table 4: Within-run accuracy and precision results of ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin at LLOQ (10 ng/mL), (n = 6)

ACCEPTED MANUSCRIPT STDV

CV%

Accuracy %

Ciprofloxacin

10.85

0.77

7.77

108.50

Gatifloxacin

10.28

0.49

5.20

102.83

Moxifloxacin

10.18

0.57

6.10

101.83

9.63

0.16

1.82

Ofloxacin

T

Mean

SC RI P

Analyte

96.33

NU

CV: Coefficient of Variation, n: Number of Samples, STDV: Standard Deviation.

Analyte

Mean 9.52

Gatifloxacin

9.73

PT

ED

Ciprofloxacin

9.98

CE

Moxifloxacin

10.14

STDV

CV%

Accuracy %

1.33

14.40

95.18

0.83

8.74

97.28

0.58

6.00

99.78

0.90

9.12

101.39

AC

Ofloxacin

MA

Table 5: Between-run accuracy and precision results of ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin at LLOQ (10 ng/mL), (n = 6)

CV: Coefficient of Variation, n: Number of Samples, STDV: Standard Deviation.

Table 6: IS-normalized matrix factor results for ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin at Low QC and High QC (10 ng/mL), (n = 6)

ACCEPTED MANUSCRIPT CV: Coefficient of Variation, n: Number of Samples, QC: Quality Control, STDV: Standard Deviation.

Gatifloxacin

High QC

Mean

1.185

1.132

STDV

0.086

0.08

CV%

7.22

0.978

0.955

STDV

0.059

0.049 5.18

0.998

1.002

0.044

0.065

4.39

6.53

Mean

0.788

0.796

STDV

0.085

0.061

CV%

10.82

7.72

ED

STDV

MA

6.05

Mean

CE

PT

CV%

AC

Ofloxacin

7.08

Mean

CV%

Moxifloxacin

T

Low QC

SC RI P

Ciprofloxacin

Criteria

NU

Analyte

Table 7: Short-term stability results for ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin at Low QC and High QC (10 ng/mL), (n = 6) Analyte Ciprofloxacin

Criteria

Low QC

High QC

Mean

26.617

922.383

ACCEPTED MANUSCRIPT Low QC

High QC

STDV

2.045

45.580

CV%

7.68

4.94

Accuracy %

88.72

108.52

Mean

27.433

STDV

2.007

52.160

CV%

7.31

5.76

106.50

26.533

917.517

1.721

45.366

6.49

4.94

88.44

107.94

Mean

26.633

873.317

STDV

1.979

52.696

CV%

7.43

6.03

88.78

102.74

Mean

ED

STDV

PT

CV%

AC

CE

Accuracy %

Ofloxacin

905.267

91.44

MA

Accuracy %

NU

Gatifloxacin

Moxifloxacin

T

Criteria

SC RI P

Analyte

Accuracy %

CV: Coefficient of Variation, n: Number of Samples, QC: Quality Control, STDV: Standard Deviation.

ACCEPTED MANUSCRIPT Table 8: Freeze-thaw stability results for ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin at Low QC and High QC (10 ng/mL), (n = 6) Criteria

Low QC

High QC

Mean

27.733

890.717

STDV

1.827

45.412

CV%

6.59

Accuracy %

92.44

104.79

Mean

28.150

914.083

NU

39.504

6.99

4.32

Accuracy %

93.83

107.54

Mean

26.217

846.700

STDV

1.393

43.318

CV%

5.32

5.12

Accuracy %

87.39

99.61

Mean

27.067

886.633

STDV

1.648

35.341

CV%

6.09

3.99

Accuracy %

90.22

104.31

Gatifloxacin

CE

PT

ED

CV%

AC

5.10

1.966

MA

STDV

Moxifloxacin

SC RI P

Ciprofloxacin

T

Analyte

Ofloxacin

ACCEPTED MANUSCRIPT Analyte

Criteria

Low QC

High QC

CV: Coefficient of Variation, n: Number of Samples, QC: Quality Control, STDV: Standard Deviation.

Concentration

SC RI P

T

Table 9: Mean, minimum and maximum concentration (ng/mL) of ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin in human aqueous humor samples Ciprofloxacin Gatifloxacin Moxifloxacin Ofloxacin

69.52

Minimum Concentration

11.2

MA

Mean Concentration

178.7

(n=27)

(n=11)

271.70

891.79

191.42

94.6

131.1

50.4

692.1

1920.2

428.5

CE

PT

ED

Maximum Concentration

(n=16)

NU

(n=13)

AC

Table 10: Median minimum inhibitory concentrations (ng/mL) of bacterial endophthalmitis isolates to ciprofloxacin, gatifloxacin, moxifloxacin and ofloxacin . Reproduced with permission from (Mather, et al., 2002). Pathogen

Ciprofloxacin Gatifloxacin Moxifloxacin Ofloxacin

Susceptible Staphylococcus aureus

320

110

60

630

Streptococcus

750

220

90

2000

ACCEPTED MANUSCRIPT pneumonia Enterococci species

750

380

190

2000

Bacillus species

130

90

90

380

AC

CE

PT

ED

MA

NU

SC RI P

T

Figure 1: Chromatograms for selectivity test of 5 different blank aqueous humor samples (a), (b), (c), (d) and (e), and chromatograms of aqueous humor samples at LLOQ level (10 ng/mL) of (f) ciprofloxacin, (g) gatifloxacin, (h) moxifloxacin, (i) ofloxacin and (j) gemifloxacin.

AC

CE

PT

ED

MA

NU

SC RI P

T

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

NU

SC RI P

T

Figure 2: Mean concentrations of moxifloxacin, gatifloxacin, ofloxacinand ciprofloxacin in patients’ aqueous humor at 2 time points:

AC

CE

PT

ED

MA

Figure 3: Comparison of the 4 fluoroquinolones mean concentrations in patients’ aqueous humor to the MIC levels of the most common pathogens causing endophthalmitis: