Adoption of the New CLSI Fluoroquinolone Breakpoints for Enterobacteriaceae

LETTER

We read with interest the minireview titled “Don’t Get Wound Up: Revised Fluoroquinolone Breakpoints for Enterobacteriaceae and Pseudomonas aeruginosa” by Van et al. (1). It provided a useful perspective on the practical implementation of CLSI breakpoint revisions, and we would like to add to the discussion by sharing our recent experience.

The ciprofloxacin and levofloxacin breakpoints for Enterobacteriaceae were revised in the CLSI document M100, 29th edition (2), released in January 2019. However, the breakpoints for nalidixic acid and other fluoroquinolone antibiotics have remained unchanged. As antimicrobial susceptibility testing (AST) results for both levofloxacin and nalidixic acid for urinary tract isolates were reported routinely by our laboratory, we decided to examine the impact of the revised fluoroquinolone breakpoints on our AST result reporting.

The MIC values and susceptibility categories for levofloxacin and nalidixic acid among Enterobacteriaceae isolated from urinary tract specimens by our laboratory in 2018 are shown in Tables 1 and 2, respectively. In total, susceptibility testing was performed for 10,115 Enterobacteriaceae urinary tract isolates in 2018. The revised levofloxacin breakpoints resulted in significant increase in levofloxacin-intermediate and -resistant isolates from 330 (3.3%) to 2,566 (25.4%) and from 2,471 (24.4%) to 2,862 (28.3%), respectively, which is in line with an expected increase of 4% in the proportion of resistant isolates (3). Notably, the number of nalidixic acid-susceptible isolates which were levofloxacin intermediate or resistant increased sharply from 26 (0.3%) to 607 (6.0%) and 8 (0.1%) to 50 (0.5%), respectively.

TABLE 1.

Distribution of MIC values for levofloxacin and nalidixic acid among Enterobacteriaceae isolated from urinary tract specimens in 2018

Nalidixic acid MIC (mg/liter)	No. of isolates by levofloxacin MIC (mg/liter) of:
	≤0.12	0.25	0.5	1	2	4	≥8
≤2	3,284	8	16	34	0	2	0
4	758	9	20	151	0	2	6
8	60	6	6	220	1	0	0
16	17	3	15	202	15	22	2
≥32	81	78	326	1,959	45	304	2,463

TABLE 2.

Susceptibility categories for levofloxacin and nalidixic acid among Enterobacteriaceae isolated from urinary tract specimens in 2018

Levofloxacin/nalidixic acid susceptibility resultsa	No. (%) of isolates by M100 interpretive criteria for:
	28th ed (2018)	29th ed (2019)
S/S	4,825 (47.7)	4,202 (41.5)
S/R	2,489 (24.6)	485 (4.8)
I/S	26 (0.3)	607 (6.0)
I/R	304 (3.0)	1,959 (19.4)
R/S	8 (0.1)	50 (0.5)
R/R	2,463 (24.3)	2,812 (27.8)

^aS, susceptible; I, intermediate; R, resistant.

While CLSI has adequately explained the rationale for the revised fluoroquinolone breakpoints for Enterobacteriaceae (3), it is less clear how the unrevised nalidixic acid and other fluoroquinolone breakpoints should be interpreted, especially alongside the revised ciprofloxacin and levofloxacin breakpoints. Given the similarities in pharmacokinetics/pharmacodynamics and mechanisms of resistance among these agents, the sudden increase in the number of discordant results would cause uncertainty in clinical interpretation. In our case, there is particular concern as to whether nalidixic acid remains a reliable treatment option for urinary tract infections caused by nalidixic acid-susceptible organisms which are levofloxacin intermediate or resistant. After careful consideration, our laboratory adopted the new fluoroquinolone breakpoints for Enterobacteriaceae but stopped reporting nalidixic acid AST results to avoid potential confusion. We suggest that other clinical laboratories also examine the impact on discordant AST results among the same antimicrobial class before adopting revised breakpoints.