LEllERS TO THE EDITORS

In Vivo Selection of Streptococcus pneumoniae, Resistant to Quinolones, Including Sparfloxin Louis Bernard‘, Jean-Claude Nguyen ?An2 andlean-Luc Mainardi’

Pneumococcal infections are an important cause of morbidity and mortality. The development of resistance to p-lactams has restricted the choice of antibiotics for therapy (1). Pneumococcal strains are often resistant to other antibiotics, emphasizing the need for new effective compounds. New fluoroquinolones such as sparfloxacin, the most active quinolone against Streptococcus pneumoniae (MICgo= 0.5 p,g/ml) (2), are considered as alternatives for the treatment of res- piratory infections. However, 0.5% of S. pneumoniae in France are resistant to sparfloxacin ( M I 0 4 pg/ml) and other fluoroquinolones (3, 4). The selection of S. pneumoniae resistant to quinolones has also been demonstrated in vitro (5). We report herein the in vivo selection of a strain of S. pneumoniae resistant to fluoro- quinolones, including sparfloxacin.

In November 1994, a 31-year-old HIV-infected patient, allergic to penicillin, was hospitalized for the treatment of a bronchitis and a maxillary sinusitis due to Haemophilus injuenzae and S. pneumoniae serotype 14, resistant to macrolides and co-trimoxazole but susceptible to penicillin (MIC = 0.015 pg/ml), tetra- cycline and chloramphenicol. The MICs of pefloxacin, ofloxacin, ciprofloxacin and sparfloxacin were 32, 4, 4 and 0.5 pg/ml respectively. Pefloxacin (400 mg bid i.v.) was administered empirically for 48 h, with- out improvement, and replaced by ofloxacin (200 mg bid orally) for 10 days. Clinical recovery was obtained. Ten days after the last administration of ofloxacin, a therapeutic regimen including clarithromycin, lamprene, myambutol and ciprofloxacin (500 mg tid) was started because one blood culture was positive with

Corresponding author and reprint requests: Dr. Jean-Luc Mainardi. Tel: 33 1 44 12 34 53 Fax: 33 1 44 12 32 34

’ Fondation HBpital Saint-Joseph, Laboratoire de Microbiologie MBdicale, 185 rue Raymond Losserand, 75674 Paris Cedex 14, France.

* HBpital Broussais, Laboratorie de Microbiologie Mldicale, 96 rue Didot, 75674 Paris Cedex 14, France.

Mycobacterium avium. At the same time, S. pneumoniae was isolated from the sputum. The strain had the same pulsotype in pulse field gel electrophoresis and the same serotype 14 as the previous one and showed the same pattern of antibiotic resistance except for the quinolones. MICs were now: pefloxacin >128 p,g/ml, ofloxacin 32 p,g/ml, ciprofloxacin 32 pg/ml and spar- floxacin 16 pg/ml. Treatment was postponed because the patient had no clinical symptoms. One week later, he was rehospitalized for a bilateral pneumonia, with positive blood culture for the same strain of S. pneumoniae. Recovery was obtained with the administration of teicoplanin (400 mg bid i.m. for 48 h and then 400 mg daily for 10 days). The original isolate probably had a dissociated low-level resistance to the quinolones, undetectable by the usual methods.

In conclusion, the first treatment with ofloxacin was clinically successhl; a high dose of ciprofloxacin was given subsequently for a Mycobacterium avium infection, and a variant highly resistant to the quino- lones was selected and became responsible for a severe relapse. Our comments are in fact questions: “How should we interpret a dissociated susceptibility to fluoroquinolones?”. Was the higher MIC to pefloxacin an indication of low-level resistance to other agents? Would the use of sparfloxacin as initial therapy have been justified and prevented the emergence of the resistant variant?

References

1. Klugman KP. Pneumococcal resistance to antibiotics. Clinical Microbiology Review 1990; 3: 171-196.

2. Simor AE. Comparative in vitm activities of spadoxacin (CI-978; AT-4140 and other antimicrobial agents against staphylococci, enterococci, and respiratory tract pathogens. Antimicrob Agents Chemother 1990; 11: 2283-2286.

3. Geslin F’, Frkmaux A. Epidtmiologie de la rtsistance aux antibiotiques de Streptococcus pneumoniae en France (1 984- 1993). Mtdecine et Maladie Infectieuse 1994; 24: 948- 961.

4. Kitzis MD, Miegi M, AcarJE Selection de mutants resistants de Streptococcus pneumoniae par difftrentes fluoroquinolones.

60

Bernard, In Vivo Selection o f Streptococcus Pneumoniae, Resistant t o Quinolones 6 1

Abstract no. 188. 14 Reunion interdisciplinaire de chimio- thtrapie anti-infectieuse (RICAI), 1994.

5. Lafredo SC, Foleno BI3. Induction of resistance of Strepto- mccirspneirmoniae to quinolones in vitrci. Chemotherapy 1993; 39: 36-39.

Reply by Professor Richard Wise

The authors pose a number of interesting questions which revolve round the observation that, apparently, a relatively greater degree of insusceptibility can occur to one or more quinolones in comparison with others. What is the clinical significance of this observation?

The increases in the MICs between the two isolations of S. pneurnoniae were >%fold for pefloxacin, %fold for ofloxacin and ciprofloxacin and 32-fold for sparfloxacin. The original isolate must be considered to be more resistant to all these agents than the majority of strains, as the MIC90 values reported are 8->16 p g / d of pefloxacin, 1-2 p g / d of ofloxacin and ciprofloxacin (l), and 0.25-0.5 p g / d of sparfloxacin (2). Hence the questions to answer are: what were the mechanisms of resistance in the original isolate and in the later isolate, and why was the susceptibility to sparfloxacin more affected than that to the other quinolones?

Alterations to DNA gyrase in gram-positive and gram-negative bacteria is usually associated with a 4- to 32-fold increase in MIC of ciprofloxacin (3) and a parallel increase is seen for other quinolones. Alterations to the inner membrane in gram-positive bacteria can occur. In Staphylococcus aureus the norA

Prof. Richard Wise, Microbiology Department, City Hospital Trust, Dudley Road, Birmingham 818 7QH. UK. Tel: 44 121 554 3801 Ext 4255 Fax: 44 121 551 7763

gene can alter drug efRux mechanisms, and this can differentially affect the various quinolones (3). Indirect evidence from our own investigations suggests that sparfloxacin has an enhanced capacity to pass certain mammalian membranes, and therefore, possibly bacterial membranes. Hence the possession of both a norA gene and a gyrA mutation might well be expected to affect susceptibility to sparfloxacin more than to other quino- lones. However, this suggestion is just speculation and requires confirmation.

To answer the other points . . . to use the most active quinolone (sparfloxacin) at full dosage would certainly seem to carry the greatest likelihood of a successful outcome. Ofloxacin was used at a rather low dose in this case, in my opinion. The problem ofwhich quinolone to use to test the susceptibility to the class of quinolones is a thorny issue. The Working Party of the British Society for Antimicrobial Chemo- therapy (4) suggested that the least active member of a group of similar agents should be chosen to represent the group, because “false sensitivity” rather than “false resistance” is the lesser of two evils in susceptibility reporting.

References

1. Gonzalez JP, Henwood JM. Pefloxacin: a review of its antibacterial activity. Drugs 1989; 37: 628-668.

2. Cooper MA, Andrews JM, Ashby JP, Matthews RS, Wise R. In-vitro activity of sparfloxacin, a new quinolone antimicrobial agent. J Antimicrob Chemother 1990; 26: 667-676.

3. Hooper D, Wolfson JS. Mechanism of bacterial resistance to quinolones. In Qitinolone Antimicrobials (ed Hooper D, Wolfson JS) 1993 Amer SOC Microbiol, Washington.

4. Working Party of the British Society for Antimicrobial Chemotherapy. A Guide to Sensitivity Testing. J Antimicrob Chemother; 27: Suppl. D.