Prevalence and Natural History of Colonization With Fluoroquinolone-Resistant Gram-Negative Bacilli in Community-Dwelling People With Spinal Cord Dysfunction

Article Outline

• Abstract
• Methods
  • Study Design
  • Study Population
  • Study Variables
  • Study Analysis
• Results
• Discussion
  • Summary of Main Results Placed in Context of Other Studies
  • Strengths and Weakness
• Conclusions
• Acknowledgments
• References
• Copyright

Abstract 

Roghmann M-C, Wallin MT, Gorman PH, Johnson JA. Prevalence and natural history of colonization with fluoroquinolone-resistant gram-negative bacilli in community-dwelling people with spinal cord dysfunction.

Objectives

To estimate the prevalence of colonization with fluoroquinolone-resistant gram-negative bacilli (GNB) and to assess risk factors for and define the natural history of colonization.

Design

Cross-sectional study with follow-up of up to 1 year.

Setting

Multiple outpatient sites.

Participants

Eighty-four community-dwelling adults with spinal cord dysfunction.

Interventions

Not applicable.

Main Outcome Measure

Colonization of perineum with fluoroquinolone-resistant GNB.

Results

Overall, 24% of this community-dwelling cohort was colonized with fluoroquinolone-resistant GNB with resistance to both levofloxacin and gatifloxacin. The use of any antibiotic in the last 90 days was most strongly associated with an increased risk of colonization with fluoroquinolone-resistant GNB (relative risk, 2.3; 95% confidence interval, 1.1−4.8; P=.05). Incontinence was significantly more common among those with fluoroquinolone-resistant GNB colonization. Among the participants with incontinence (n=42), the use of intermittent catheterization (vs suprapubic, urethral, or condom catheters or overflow incontinence) was significantly protective. Colonization was not associated with an increase in urinary tract infection and was often transient.

Conclusions

Fluoroquinolone resistance among GNB colonizing community-dwelling people with spinal cord dysfunction is common. Colonization is more common among those receiving antibiotics and less common among those continent of urine. Among those with incontinence, those using intermittent catheterization have less colonization.

Key Words: Antibiotic resistance, Anti-infective agents, fluoroquinolone, Cross-sectional study, Multiple sclerosis, Rehabilitation, Spinal cord injuries, Urinary tract infections

ANTIBIOTIC RESISTANCE IS A growing problem in gram-negative bacteria such as Escherichia coli, Klebsiella species, and Pseudomonas aeruginosa. Resistance to fluoroquinolone antibiotics in P. aeruginosa has increased by 37% in intensive care units with 33% of P. aeruginosa resistant to fluoroquinolones in 2002.¹ Resistance to fluoroquinolones in E. coli and K. pneumoniae has also been increasing in both inpatients and outpatients.²

Colonization and infection with antibiotic-resistant bacteria has long been a major problem for people with spinal cord injury (SCI) and disease (SCD)³, ⁴, ⁵; however, many of the studies in this population have been conducted in the inpatient setting.³, ⁶, ⁷ There have been a few reports of antibiotic resistance in a community setting. Waites et al⁸ reported resistance to 2 or more antibacterial agent categories in 33% of bacterial isolates from sterilely collected urine samples of outpatients with SCI returning for annual checkups. Hinkel et al⁴ reported an increasing number of fluoroquinolone-resistant gram-negative bacilli (GNB) obtained by aseptic catheterization among outpatients with SCI at their institution.⁴ GNB commonly cause urinary tract infections (UTIs), which are the most frequent infection in people with SCI.³ Antibiotic resistance can lead to inappropriate antibiotic selections, which can in turn lead to treatment failures.⁹ In addition, parenteral antibiotics are needed when UTIs are caused by bacteria resistant to all oral antibiotics.

Because most UTIs are caused by endogenous gastrointestinal flora, understanding the risk factors for colonization with antibiotic-resistant GNB in community-dwelling people with SCI or SCD can improve our care of these patients. We focused on resistance to fluoroquinolones because this class of antibiotics is typically the most active oral antibiotic class available for the treatment of UTIs. The objective of this study was to estimate the prevalence of colonization with fluoroquinolone-resistant GNB, to assess risk factors for colonization, and to define the natural history of colonization in a population of community-dwelling people with SCI or SCD.

Back to Article Outline

Methods 

Study Design 

This was a prospective study of community-dwelling adults with SCI or SCD (hereafter referred to collectively as SCD). We cultured the perineum of participants for fluoroquinolone-resistant GNB and assessed risk factors for colonization in a cross-sectional study. We then followed participants for up to 1 year for antibiotic use and UTI. If participants were previously colonized, we repeated their cultures to determine if colonization was persistent.

Study Population 

From April 2003 to June 2004, participants were recruited from multiple outpatient sites including a university-affiliated rehabilitation center (Kernan Orthopaedics and Rehabilitation Hospital, Baltimore, MD), 2 Veterans Affairs medical centers (VAMCs) (Baltimore, MD; Washington, DC), and a university-affiliated center for multiple sclerosis (MS) (Maryland Center for Multiple Sclerosis, Baltimore, MD). We used patient support groups (Paralyzed Veterans of America, Multiple Sclerosis Society) to increase study exposure during our recruitment period.

Adults with an SCD who have lived in the community more than 90% of the past 3 months were eligible to participate. SCD was defined as chronic (>3mo) complete or incomplete motor or sensory dysfunction because of either a posttraumatic or a medical disorder of the spinal cord. SCD from nontraumatic SCDs could include but were not limited to MS, spondylotic myelopathy, syringomyelia, epidural abscess or hemorrhage, spinal cord tumor, virus-related transverse myelitis, spinal cord infarction, or other familial or degenerative spinal cord disease. People with MS and associated urinary dysfunction were eligible to participate. Urinary dysfunction in people with MS was defined as any or all of the following urinary tract symptoms in the last 12 months: frequency, nocturia, hesitancy, retention, urgency, incontinence, dysuria, or change in urine color or odor.¹⁰

Participants provided informed consent and were followed longitudinally for up to 1 year. The institutional review boards of the University of Maryland, Baltimore, and the Washington, DC, VAMC approved the study protocol. Implementation of this study conformed to the approved protocol, as well as the ethical and humane principles of research.

Study Variables 

We collected the following at enrollment from participants and their medical records: age, sex, ethnic background, etiology of SCD, years since onset of SCD, level of disability as measured by the Eastern Cooperative Oncology Group (ECOG),¹¹ presence of decubiti, method of emptying bladder, and previous hospitalizations and antibiotics within the last 90 days. If participants had MS, we classified them by subtype.¹² If participants had SCI because of trauma, we collected their American Spinal Injury Association (ASIA) Impairment Scale score.¹³ We chose to categorize the 6-point ECOG score as 0 to 2 and 3 to 5. People with an ECOG score of 0 to 2 are capable of all self-care. People with an ECOG score of 3 to 5 are capable of only limited self-care or completely disabled.

We obtained cultures of the perineal skin on enrollment and then quarterly. Cultures were plated on MacConkey agar and MacConkey agar with 4μg/mL of ciprofloxacin to screen for the presence of fluoroquinolone-resistant GNB in a research microbiology laboratory. Ciprofloxacin-resistant bacteria were identified by using Vitek GNI cards or API20E.^a We used Etest assays^b to determine susceptibility to trimethoprim-sulfamethoxazole, gatifloxacin, levofloxacin, ertapenem, imipenem, ampicillin and sulbactam, piperacillin and tazobactam, and ceftazidime per Clinical and Laboratory Standards Institute standards.¹⁴ We collected data on antibiotic use including days on intravenous and oral (as applicable) antibiotics during the time in which the patients were being followed from their medical records. All patients were followed for at least 3 months. In addition, we determined whether the antibiotic was prescribed for a UTI.

Urine for culture was collected via sterile intermittent catheterization when clinically indicated. UTIs were defined as 1 of the following: a urine culture of 10⁵ colony-forming units (cfu) per milliliter or more in a patient with at least 1 symptom of UTI or at least 2 urine cultures of 10³cfu/mL or more in a patient with at least 2 of the common symptoms of UTI. Common symptoms of UTI included the following: fever (>38°C), urgency, frequency, dysuria, and suprapubic tenderness. Autonomic dysreflexia (hypertension, headache, visual changes, altered heart rate, flushing, diaphoresis, and/or nasal congestion) was also considered a symptom of UTI in our SCI participants. Worsening of MS symptoms was also considered a symptom of a UTI in patients with MS. The clinical microbiology laboratories for these outpatient sites worked up urine cultures from patients with SCI for 1 or more bacteria at concentrations of 10³cfu/mL or more.

Study Analysis 

Summary statistics included means and standard deviations for continuous variables and proportions for categorical variables. Group means were compared by using the t test or 1-way analysis of variance and proportions by the Pearson chi-square test or Fisher exact test, as appropriate. All statistical analyses were performed by using SPSS^c statistical software.

Back to Article Outline

Results 

Table 1 gives a description of our participants. Of note, 83% were men. Fifty-two percent had MS, and 46% had trauma as a cause of their SCD. Of the 44 participants with MS, 20 were classified as relapsing remitting, 19 as secondary progressive, 3 as primary progressive, and 2 were not specified. Of the 36 participants with SCI because of trauma, 21 were ASIA grade A, 4 were B, 7 were C, and 4 were D. Of the remaining 4 participants, 1 had Brown-Sequard syndrome, 2 had central cord syndrome, and 1 had cauda equine syndrome.

Table 1. Description of Community-Dwelling People With SCD (N=84)

Variables	Values
Demographics
Male (%)	83
Mean age ± SD (y)	50±10
Ethnic background (%)
White, non-Hispanic	65
Black, non-Hispanic	32
Other	2
Characteristics of SCD
Etiology of SCD (%)
Trauma	45
MS	52
Other	2
Mean years with SCD ± SD	17±12
ECOG score of 3−5 (%)	17
Presence of pelvic decubiti (%)	7
Method of emptying bladder (%)
Continent	50
Condom catheter	14
Intermittent catheterization	18
Suprapubic catheter	4
Urethral catheter	7
Incontinent (no catheter use)	7
Potential risk factors (%)
Hospitalizations >48h in last 90d	6
UTI in past 90d	6
Antibiotics in last 90d	26
Fluoroquinolone antibiotics in last 90 days	12
Outcome variable (%)
Colonization with fluoroquinolone-resistant GNB	24
Outcomes during follow-up
Mean days followed ± SD	306±188
UTI requiring antibiotic treatment during follow-up (%)	10
UTI because of fluoroquinolone-resistant GNB during follow-up (%)	7

Abbreviation: SD, standard deviation.

Half of the cohort was continent, 43% used some type of urinary catheter, and 7% had overflow incontinence. Twenty-seven percent had used antibiotics in the 90 days before study enrollment, and 6% were hospitalized in the 90 days before enrollment.

Overall, 24% of this community-dwelling cohort was colonized with fluoroquinolone-resistant GNB with resistance to both levofloxacin and gatifloxacin. All levofloxacin-resistant isolates were also gatifloxacin resistant and vice versa. Pseudomonas species were the most common fluoroquinolone-resistant GNB (8/20), followed by Acinetobacter species (3/20), Enterobacter species (3/20), and Citrobacter, Serratia, Proteus, S. maltophilia, E. coli, and E. fergusonii (1 each). Fluoroquinolone-resistant GNB were also likely to be resistant to other antibiotics. Eighty percent were resistant to trimethoprim and sulfamethoxazole, 75% to ampicillin and sulbactam, 55% to ceftazidime, 25% to piperacillin and tazobactam, and 20% to imipenem.

Table 2 shows characteristics associated with fluoroquinolone-resistant GNB colonization. The use of any antibiotic in the last 90 days was most strongly associated with an increased risk of colonization with fluoroquinolone-resistant GNB (relative risk [RR], 2.3; 95% confidence interval [CI], 1.1−4.8; P=.05). Fluoroquinolone use alone did not increase the risk of colonization with fluoroquinolone-resistant GNB in a statistically significant degree (RR=1.31; 95% CI, 0.46−3.67; P=.56).

Table 2. Characteristics Associated With Colonization With Fluoroquinolone-Resistant GNB

Characteristics	Colonization With Fluoroquinolone-Resistant GNB
	Yes (n=20)	No (n=64)	P
Demographics
Male (%)	85	83	.819
Mean age ± SD (y)	54±10	49±10	.916
Ethnic background (%)			.313
White, non-Hispanic	75	63
Black, non-Hispanic	20	36
Other	5	2
Characteristics of SCD
Etiology of SCD (%)			.026
Trauma	50	44
MS	40	56
Other	10	0
Mean years since SCD ± SD	14	11	.096
ECOG 3−5 (%)	25	14	.252
Presence of pelvic decubiti (%)	10	6	.570
Method of emptying bladder (%)			.006
Continent	35	55
Condom catheter	25	11
Intermittent catheterization	0	23
Suprapubic catheter	5	3
Urethral catheter	20	3
Overflow incontinence	15	5
Potential risk factors (%)
Hospitalizations >48h in last 90d	5	6	.867
Antibiotics in last 90d	45	20	.028
Fluoroquinolone antibiotics in last 90d	15	11	.634
Outcomes during follow-up
UTI during follow-up (%)	10	9	.934
UTI because of fluoroquinolone-resistant GNB during follow-up (%)	10	6	.570
Mean days followed ± SD	210	182	.334

Method of emptying the bladder was associated with colonization with fluoroquinolone-resistant GNB. Incontinence was more common among those with fluoroquinolone-resistant GNB colonization. Prior antibiotic use was also strongly associated with incontinence (50% vs 2%; χ² test, P<.001). Among the participants with incontinence (n=42), the use of intermittent catheterization (vs suprapubic, urethral, or condom catheters or overflow incontinence) was significantly protective. None of 15 patients using intermittent catheterization were colonized compared with 44% of 27 patients using other methods of bladder management (0% vs 44%; Fisher exact test, P=.002). Among the participants with incontinence, those using intermittent catheterization were not less likely to have had antibiotics in the prior 90 days than those using other methods (40% vs 55%; Fisher exact test, P=.33).

The duration of follow-up did not differ significantly in those with colonization compared with those without colonization. There were no significant differences in UTI during follow-up or UTI caused by fluoroquinolone-resistant GNB during follow-up. We obtained follow-up cultures on 16 of the 20 people colonized with fluoroquinolone-resistant GNB. Four of the16 people (25%) had persistent colonization on their final culture; neither antibiotic use nor days of follow-up were associated with persistence (data not shown). People with fluoroquinolone resistant P. aeruginosa were more likely to have persistent colonization compared with other fluoroquinolone-resistant GNB (75% vs 16%, Fisher exact test, P=.06).

Back to Article Outline

Discussion 

Summary of Main Results Placed in Context of Other Studies 

We found that 24% of community-dwelling patients with SCD were colonized with fluoroquinolone-resistant GNB, which is a high prevalence of resistance in an outpatient setting. In contrast, Waites et al⁸ found that 9% of Enterobacteriaceae species were resistant to ciprofloxacin; however, their data were collected from 1993 to 1997. Hinkle et al⁴ reported increasing proportions of E. coli, K. pneumoniae, P. mirabilis, and P. aeruginosa resistant to fluoroquinolones from 1994 to 1999, obtained from urine cultures of outpatients with SCI. Taken together, this suggests that resistance to fluoroquinolones in the community setting has increased considerably in the past decade.

We also found that the recent use of any antibiotic (not fluoroquinolone antibiotics alone) was the strongest risk factor for antibiotic-resistant colonization. This contrasts with the findings of Gimber et al¹⁵ who reported that the use of fluoroquinolone antibiotics (not any antibiotic) was strongly associated with ciprofloxacin-resistant E. coli bacteriuria when compared with patients with ciprofloxacin-susceptible E. coli bacteriuria. Although fluoroquinolone use clearly can select for resistance,¹⁶ this difference is likely because of methodologic differences in control-group selection (noncolonized individuals in our study vs those with bacteriuria because of susceptible organisms in the Gimber study) and the culturing method (surveillance cultures in our study vs clinical cultures in the Gimber study). The use of antibiotics other than fluoroquinolones may also select for fluoroquinolone resistance because of the presence of resistance to multiple classes of antibiotics in GNB resistant to fluoroquinolones. For example, most of the colonizing fluoroquinolone-resistant GNB were also resistant to trimethoprim and sulfamethoxazole.

Continent patients were less likely to be colonized with fluoroquinolone-resistant GNB. This is possibly because they were less likely to have recently used antibiotics. These patients were also not as frequently instrumented with a catheter, which might also have played a role in this finding. Interestingly, among those with incontinence, people who used intermittent catheterization were less likely to have fluoroquinolone-resistant GNB. The intermittent catheterization group, however, was not significantly different in recent antibiotic use from the incontinent group not using intermittent catheterization. Waites et al⁸ also found that persons with indwelling catheters or condom catheters were at higher risk for bacteriuria with a bacterial isolate resistant to 2 or more antibiotic classes. It is possible that the presence of a permanent urologic device such as a Foley or condom catheter increases the risk of colonization with resistant organisms through the promotion of persistent colonization.

Finally, we did not find differences in the rate of UTIs between colonized and not-colonized participants. We also found that colonization was often transient with the exception of colonization with fluoroquinolone-resistant P. aeruginosa. This suggests that clinicians may need to consider alternative antibiotics over the long term once colonization with this organism has been detected.

Strengths and Weakness 

Our study has a number of important strengths. We studied an outpatient population recruited from multiple diverse sites. Few studies have looked at antimicrobial resistance in a community-dwelling population of people with SCI,⁴, ⁸, ¹⁵ and even fewer have assessed risk factors for resistance.⁸, ¹⁵ Although antibiotic resistance has long been recognized as an issue for people with SCI, most studies have been performed in the inpatient setting in which patient-to-patient transmission was likely an important contributor to the prevalence of resistance. We also used surveillance cultures, not clinical cultures (which are taken in response to the clinical suspicion of an infection), to measure the presence of resistance and collected our data prospectively, thereby optimizing the accuracy of our data.

Our study was limited by a relatively small sample size and loss to follow-up; however, our loss to follow-up was distributed evenly between those with and without colonization. Despite this, we suspect that our sample size was not sufficient to detect a difference in UTI if one existed because of the small sample size and the low frequency of UTI.

Back to Article Outline

Conclusions 

We found that colonization with fluoroquinolone-resistant GNB was common in a community-dwelling population of people with SCD. Those colonized with fluoroquinolone-resistant GNB were more likely to have used antibiotics recently and were less likely to be continent. Among those who were not continent, people who used intermittent catheterization were less likely to be colonized with fluoroquinolone-resistant GNB. Given the high prevalence of resistance, treatment of infections in this population needs to be guided by cultures; however, given the risk of resistance from antibiotic use, we need to be careful to treat only symptomatic infections, not asymptomatic colonization or contamination. The use of intermittent catheterization appears to be protective; practitioners should encourage its use by incontinent people with SCD.

Suppliers

Back to Article Outline

Acknowledgments 

We thank Debbie Grady, RN, for her critical role in coordinating the project. We thank the clinical staff of enrollment sites and other members of the research team including the VA MS Center of Excellence-East. We also thank Ebbing Lautenbach, MD, MPH, for his critical reading of the manuscript and helpful suggestions.

Back to Article Outline

References 

1. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2003, issued August 2003. Am J Infect Control. 2003;31:481–498
   • View In Article
   • Full Text
   • Full-Text PDF (218 KB)
   • CrossRef
2. Lautenbach E, Strom BL, Nachamkin I, et al. Longitudinal trends in fluoroquinolone resistance among Enterobacteriaceae isolates from inpatients and outpatients, 1989–2000: differences in the emergence and epidemiology of resistance across organisms. Clin Infect Dis. 2004;38:655–662
   • View In Article
   • CrossRef
3. Montgomerie JZ. Infections in patients with spinal cord injuries. Clin Infect Dis. 1997;25:1285–1290
   • View In Article
   • MEDLINE
4. Hinkel A, Finke W, Botel U, Gatermann SG, Pannek J. Increasing resistance against antibiotics in bacteria isolated from the lower urinary tract of an outpatient population of spinal cord injury patients. Urol Int. 2004;73:143–148
   • View In Article
   • MEDLINE
   • CrossRef
5. Penders J, Huylenbroeck AA, Everaert K, Van Laere M, Verschraegen GL. Urinary infections in patients with spinal cord injury. Spinal Cord. 2003;41:549–552
   • View In Article
   • MEDLINE
   • CrossRef
6. Mylotte JM, Kahler L, Graham R, Young L, Goodnough S. Prospective surveillance for antibiotic-resistant organisms in patients with spinal cord injury admitted to an acute rehabilitation unit. Am J Infect Control. 2000;28:291–297
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (41 KB)
7. Mylotte JM, Graham R, Kahler L, Young L, Goodnough S. Epidemiology of nosocomial infection and resistant organisms in patients admitted for the first time to an acute rehabilitation unit. Clin Infect Dis. 2000;30:425–432
   • View In Article
   • MEDLINE
   • CrossRef
8. Waites KB, Chen Y, DeVivo MJ, Canupp KC, Moser SA. Antimicrobial resistance in gram-negative bacteria isolated from the urinary tract in community-residing persons with spinal cord injury. Arch Phys Med Rehabil. 2000;81:764–769
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (160 KB)
   • CrossRef
9. Dow G, Rao P, Harding G, et al. A prospective, randomized trial of 3 or 14 days of ciprofloxacin treatment for acute urinary tract infection in patients with spinal cord injury. Clin Infect Dis. 2004;39:658–664
   • View In Article
   • CrossRef
10. Multiple Sclerosis Council for Clinical Practice Guidelines (Urinary dysfunction and multiple sclerosis). New York: National Multiple Sclerosis Society; 1999;
    • View In Article
11. Oken MM, Creech RH, Tormey DC, et al. Toxicity and response criteria of the Eastern Cooperative Oncology Group. Am J Clin Oncol. 1982;5:649–655
    • View In Article
    • MEDLINE
    • CrossRef
12. Lublin FD, Reingold SC. Defining the clinical course of multiple sclerosis: results of an international survey (National Multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of New Agents in Multiple Sclerosis). Neurology. 1996;46:907–911
    • View In Article
    • MEDLINE
13. American Spinal Injury Association, International Medical Society of Paraplegia. Standard neurological classification of spinal cord injury. Chicago: ASIA/IMSOP; 2002;
    • View In Article
14. Committee for Clinical Laboratory Standards. Performance standards for antimicrobial susceptibility testing. Wayne: National Committee for Clinical Laboratory Standards; 2006;
    • View In Article
15. Gimber EA, Shields MD, Canawati HN, et al. Bacteriuria with Escherichia coli resistant to ciprofloxacin in patients with spinal-cord injury [letter]. Infect Control Hosp Epidemiol. 1998;19:85–86
    • View In Article
    • MEDLINE
16. Waites KB, Canupp KC, Brookings ES, DeVivo MJ. Effect of oral ciprofloxacin on bacterial flora of perineum, urethra, and lower urinary tract in men with spinal cord injury. J Spinal Cord Med. 1999;22:192–198
    • View In Article
    • MEDLINE

• a bioMérieux Inc, 100 Rodolphe St, Durham, NC 27712.
• b AB Biodisk, Dalvagen 10, 169 56 Solna, Sweden.
• c Version 12.1; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.