In Vitro Antibiotic Susceptibility of Pseudomonas aeruginosa Corneal Ulcer Isolates

Pseudomonas aeruginosa is the most frequently isolated gram-negative pathogen causing bacterial keratitis in India.1 It causes a potentially devastating hypopyon keratitis, more commonly in association with the use of contact lens wear, and is one of the causes of severe necrotic ulcerative keratitis.1 Due to its rapid spread and destruction of the corneal tissue, effective treatment should be begun before any definitive evidence of antibiotic susceptibility of the causative microorganism is obtained, to salvage useful vision. The protocol for empirical treatment of bacterial keratitis involves use of fortified antibiotics. However, recent evidence supports the equivalent efficacy of fluoroquinolone monotherapy as an initial treatment.2 Knowledge of antimicrobial resistance trends in the prevalent bacterial strains is essential as it has prognostic value in the final clinical and visual outcomes of infective keratitis.3 We report antibiotic susceptibility pattern of P. aeruginosa isolates cultured from corneal scrapings of patients with infectious keratitis presenting to the outpatient department of a tertiary eye care center.

A retrospective chart review was performed for in vitro antibiotic susceptibility of P. aeruginosa isolates recovered from corneal scrapings of patients who presented with infectious keratitis to our outpatient department over the last 12 months (September 2011 – August 2012). Samples from all the patients with infectious keratitis were obtained by scraping the leading edges of the infiltrate as well as the base. This material was then inoculated directly onto culture plates for isolation and for determining the antibiotic susceptibility of the causative organism. The growth of a single colony of a non-commensal organism or at least three colonies of an opportunistic pathogen was considered to be a positive culture.4 Antibiotic susceptibility testing was performed by the standard Kirby Bauer disc diffusion technique.5 Various antibiotics (Hi-Media, Mumbai, India) used and their concentration/disc (μg) were ciprofloxacilin (10), gatifloxacin (5), moxifloxacin (5), tobramycin (10), amikacin (30), tetracycline (30), cefazolin (30), ceftazidime (30), and cefuroxime (5). Concentration of polymyxin B (Hi-Media, Mumbai, India) was 300 units/disc. Susceptibility was tested against cefazolin for pseudomonal isolates as it is routine to test the antibiotic susceptibility to the aforementioned antimicrobials for all culture-positive isolates in our laboratory irrespective of Gram status in order to give wide coverage for both Gram-positive and Gram-negative bacteria. There are no susceptibility standards of antibiotics for topical therapy in the eye, thus the systemic susceptibility standards were used, based on the assumption that greater or equal concentrations of drugs were achieved in the ocular tissues after round the clock topical application. For testing anti-pseudomonal activity of any antibiotic, the universally accepted standards laid down by the Clinical and Laboratory Standards Institute (CLSI) for media composition, pH, disk size, lawn culture, temperature of incubation, inoculum size, etc., were strictly adhered to.6 The method adopted for polymyxin testing was the standard Kirby Bauer disc diffusion criteria, in which pseudomonal strain was considered to be resistant to polymyxin B if it showed an inhibition zone of diameter 11 mm or less around the antibiotic disc.7

Multidrug resistance (MDR) was defined as acquired nonsusceptibility to at least one agent in three or more antimicrobial categories (polymyxins, fluoroquinolones (FQ), aminoglycosides, tetracycline, and cephalosporins) against which susceptibility was tested.8 Extreme drug resistance (XDR) has been defined as ex vivo non susceptibility to at least one agent in all but two or fewer antimicrobial categories.9 Since we tested five classes of antibiotics, we considered XDR as nonsusceptibility to at least one agent in all but one antimicrobial category. The data were entered into Microsoft Excel 2010 and analyzed. Appropriate statistical analysis was done by applying nonparameteric tests: chi-square test for calculating the difference between the susceptibility of isolates to two different classes of antibiotics, and McNemar test for analyzing the association between any two classes of antibiotics. p < 0.05 was considered as statistically significant.

Of the total number of cultures performed from corneal scrapings of cases with infectious keratitis (812) between September 2011 and August 2012, 57 corneal isolates tested positive for P. aeruginosa; bacterial infections other than Pseudomonas were recovered from 487 samples; 86 showed fungal growth and 182 were culture negative (87 were sterile and 95 showed aerobic spore-forming bacteria presumed to be due to extraneous contamination). Culture positivity rate in our series was 77.6% (630/812). Pseudomonal keratitis comprised approximately one-tenth (10.5% [57/544]) of all culture-proven bacterial isolates in our series. This corresponds to previous reports that state that pseudomonal corneal ulcers constitute 15–20% of bacterial corneal ulcers.1

Mean age of patients with culture-positive pseudomonal keratitis was 42.8 ± 19.7 years (Range: 78–6 years); 42 were males and 15 were females. Of the 57 pseudomonal-positive isolates, 30 (52.6%) were MDR and 15 (26.3%) were XDR. Among the XDR, all isolates showed susceptibility to polymyxin B only. Polymyxin B showed maximum in vitro activity against the cultured organisms (100%). Among fluoroquinolones, 60% of isolates showed in vitro susceptibility to gatifloxacin, 56% to ciprofloxacin, and 51% to moxifloxacin respectively. Fifty-one percent of isolates showed susceptibility to tobramycin and 44% to amikacin among the aminoglycosides. In cephalosporin group, 39, 37, and 32% showed susceptibility toward ceftazidime, cefuroxime, and cefazolin, respectively. Thirty-five percent of isolates were susceptible to tetracycline. Matrix for association of antibiotic susceptibility between different antibiotic pairs is shown in Table 1. It shows that gatifloxacin showed significant activity toward other drug-resistant pseudomonal strains, followed by ciprofloxacin and tobramycin.

TABLE 1. Association matrix of antibiotic susceptibilities.

	Gatiflox		Moxiflox		Ciproflo		Tobra		Amika		Cefurox		Ceftazid		Cefazol		Tetracyc
	S	R	S	R	S	R	S	R	S	R	S	R	S	R	S	R	S	R
Moxiflox	8	0
R S	(0.005)
Ciproflo	6	4	2	8
R S	(0.5)		(0.06)
Tobra	8	3	1	4	9	6
R S	(0.13)		(0.18)		(0.4)
Amika	10	1	2	1	9	2	5	1
R S	(0.007)		(0.6)		(0.03)		(0.10)
Cefurox	13	0	5	0	12	1	9	1	6	2
R S	(0.0003)		(0.025)		(0.023)		(0.011)		(0.16)
Ceftazid	13	1	5	1	13	3	9	2	6	3	1	2
R S	(0.001)		(0.1)		(0.012)		(0.03)		(0.3)		(0.56)
Cefazol	16	1	12	5	20	7	14	4	13	7	11	9	11	8
R S	(0.003)		(0.09)		(0.012)		(0.02)		(0.18)		(0.65)		(0.5)
Tetracyc	14	0	7	1	12	0	11	2	7	2	5	4	6	4	8	9
R S	(0.002)		(0.03)		(0.0005)		(0.013)		(0.09)		(0.74)		(0.53)		(0.8)

p Values between any two discordant pairs are shown in brackets. Statistical test used: McNemar test; association is significant at p < 0.05 and is boldfaced. Since none of the tested pseudomonal strains showed lack of susceptibility to polymyxin B, association between polymyxin B and other antibiotics could not be evaluated. R = resistance to the tested antibiotic of the cultured pseudomonal strain, S = susceptibility to the tested antibiotic of the cultured pseudomonal strain, Moxiflox = moxifloxacin, Ciproflo = ciprofloxacin, Tobra = tobramycin, Amika = amikacin, Cefurox = cefuroxamin, Ceftazid = ceftazidime, Cefazol = cefazolin, Tetracyc = tetracycline).

Acquired nonsusceptibility to at least one of the three tested fluoroquinolones was observed in 58% of isolates, and 35% of isolates were not inhibited by any of the tested fluoroquinolones. Although susceptibility toward gatifloxacin was more than that for moxifloxacin, it was not statistically significant (p = 0.13; chi-square test). Among the ciprofloxacin-resistant Pseudomonas aeruginosa (CRPA), 24% were susceptible to gatifloxacin, whereas 8% were susceptible to moxifloxacin, a difference that is statistically significant (p = 0.02; chi-square test). This indicates that gatifloxacin might be FQ of choice against the CRPA.

Though nonocular isolates of P. aeruginosa are known to possess poly-drug-resistant phenotypes,8^,9 there are few reports on ex vivo poly-antibiotic resistance in P. aeruginosa ocular isolates.10 Often, several mechanisms act synergistically to contribute to multidrug resistance.11 These include the intrinsic resistance of organisms, a low permeability of the outer membrane to administered antibiotics, expression of chromosomal AmpC cephalosporinase; production of plasmid mediated β-lactamases against different molecular classes; and an enhanced expression of active efflux systems for wide variety of substrates. They also may synthesize aminoglycoside-modifying enzymes (phosphoryltransferases, acetyltransferases, and adenylyltransferases), an antibiotic-resistant biofilm (consisting of bacterial communities embedded in an exo-polysaccharide matrix), and cause structural alterations of topoisomerases II and IV preventing efficacy of fluoroquinolones.

Our series showed a high prevalence of pseudomonal isolates with reduced susceptibility toward FQ, which has not been reported from other parts of the world. P. aeruginosa isolates exhibited in vitro resistance ranging from 40 to 55% against FQs in our series. Further, gatifloxacin was found to inhibit a statistically greater number of ciprofloxacin-resistant pseudomonal isolates. A report from the United Kingdom has found P. aeruginosa to be 100% susceptible to moxifloxacin and 99% to ciprofloxacin.12 Mohammadpour et al. showed 100% susceptibility of Pseudomonas keratitis among contact lens wearers to ciprofloxacin and ceftazidime in Iran.13 Data from Australia showed that pseudomonal isolates from Indian subcontinent have a higher prevalence of in vitro resistance than elsewhere.14 Although studies from South India reported an incidence of 15% for ciprofloxacin-resistant Pseudomonas in 1999,10 there seems to be at least a 2-fold rise in in vitro FQ resistance over the last decade. Moreover, an intra-antibiotic class variation in susceptibility between FQs was observed.

Acquired resistance in multidrug-resistant gram-negative bacilli to polymyxin B is not common at present. This is probably due to declining use of polymyxin B over the last 50 years. P. aeruginosa is intrinsically susceptible to polymyxin B, an old class of cationic, cyclic polypeptide antibiotic. Positively charged polymyxins interact electrostatically with the negatively charged lipopolysaccharide (LPS) of the outer membrane of gram-negative bacteria and are subsequently taken up via the “self-promoted uptake” pathway leading to cell death.15 Physicians who treat systemic pseudomonal infections have reverted back to older anti-pseudomonals like colistins.

Clinical diagnosis of pseudomonal keratitis has the highest predictive value among all infectious etiologies.16 Therefore, the knowledge about the antibiotic susceptibility pattern in Pseudomonas isolates may be helpful in choosing an appropriate antibiotic to treat corneal ulcers that present with a large suppurative infiltrate along with other diagnostic signs suggestive of Pseudomonas keratitis.

The limitations of our study include that an in vitro antimicrobial susceptibility profile of cultured isolates may not truly reflect the clinical response attained in vivo. Correlation of drug resistance with clinical course may be helpful in better understanding the virulence of these strains in future studies. Since the microbiological profile and the clinical presentation differ geographically, the results from our study may not be generalized to other population groups.

In conclusion, documenting the current susceptibility pattern of the infectious strains is necessary to devise treatment protocol particular to each geographical area. According to our data, there is an emergence of in vitro multidrug resistance to most of the current topical antibiotics, including fourth-generation fluoroquinolones, in nearly 52% of P. aeruginosa keratitis isolates. These MDR strains showed 100% in vitro susceptibility toward polymyxin B. This knowledge may be useful if correlated to the clinical outcome while treating multidrug-resistant pseudomonal keratitis in Northern India.

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.