Printer Friendly

Clinical impact of antibiotic susceptibility data.

Clinical impact of antibiotic susceptibility data

Rapid laboratory results are welcomed under prospective payment because they can speed up appropriate treatment and shorten hospital stays. As the microbiology supervisor at a 32-bed rural hospital, I found more and more physicians expressing concern about the length of time required for culture and susceptibility results--often 48 hours.

Quicker service was possible. Some automated microbiology systems deliver susceptibility results within six to eight hours, but the cost is almost double that of the overnight Kirby-Bauer procedures used in our laboratory, not to mention the capital expenditure we would have to make. Before I could even consider acquiring such a system, I needed to verify that faster susceptibility results would actually influence patient treatment.

A check of the literature turned up studies with varying outcomes. One group of investigators found that 276 out of 390 patients with bacteremia, or 70.8 per cent, had empiric therapy (based on physical examination) that was altered by susceptibility results.1 In another study, culture results prompted a change in therapy for 59 per cent of patients.2 But at a Portland (Ore.) hospital, only 48 out of 173 patients with bacteremia, or 27.8 per cent, had a change of empiric therapy because of susceptibility results.3

We conducted our own study two years ago. For six months we followed 76 inpatients with positive cultures--that is, growth of bacteria considered to be significant pathogens for the particular body site (results with organisms of questionable significance were not included).

Our evaluation covered three basic areas: 1) the number of patients given empiric therapy prior to culture results, 2) how often physicians initiated antibiotic therapy or changed empiric therapy based on culture and/or susceptibility results, and 3) how the microbiology laboratory could improve procedures to get results into physicians' hands more quickly.

We send out our reports daily until the culture is completed, using a multi-layer form with NCR paper so that each report is cumulative. If a specimen arrives in the laboratory before 2 p.m., a Cram stain report is issued that same day. Gram stain and preliminary culture reports are reported within 24 hours. Interim reports are issued daily, giving as much information to physicians as possible, i.e., gram-negative rods, beta-he-molytic strep, etc. The final .lab report includes Gram stain, organism identification, and antibiotic susceptibility results wherever applicable.

One advantage of a small hospital is ease of access to patient charts. With only one nurses' station, it is easy to monitor charts. I was able to check them daily, comparing the date and time of antibiotic orders with the date and time that culture results had been delivered to the nurses' station.

It is difficult, however, if not impossible, to be totally accurate in measuring a response to laboratory data. With no way of determining what action a physician would have taken without the culture data, we assumed that any change in therapy up to 12 hours after the laboratory report was indeed influenced by that report. Since our reports often reached the floors after morning rounds, many physicians did not see them until evening; hence, the 12-hour limit in the study.

Figure I shows how the reports influenced therapy. Sixty per cent of the patients studied, or 46 out of 76, were started on antibiotics prior to any culture results. Of these, 25 remained on the original antibiotic and 21 had therapy changed after culture results were available. Eight changes were due to resistance of the isolated organism to the empiric antibiotic; 13 involved a switch to less costly or more appropriate antibiotics. Seventy-six per cent of these changes took place after susceptibility results were reported.

Among the 29 patients put on antibiotics after culture and/or susceptibility results were reported to physicians, six began therapy after Gram stain results. All were sputum specimens with smears showing S. pneumoniae or H. influenzae. Thirteen patients began therapy after a preliminary report of organism type, and the other 10 after susceptibility results. One patient had no antibiotic therapy at all noted on the chart.

We were encouraged to see that our microbiology reports influenced treatment of 50 patients (65.7 per cent of the total)--29 who had not begun on empiric therapy and 21 who changed from empiric therapy. Since antibiotic therapy was based on susceptibility reports for 26 of these patients (34 per cent), it seemed critical to obtain susceptibility results as quickly as possible. The sooner a patient is put on appropriate therapy, the sooner he or she can be discharged from the hospital.

While the study was under way, we looked into the cost of automated susceptibility systems and finally concluded that they were beyond our reach. Instead, we explored ways to speed up our manual methods.

Significant saving of time emerged from experiments with reading Kirby-Bauer plates in less than our usual 18 to 20 hours. We found that plates read at 8 to 10 hours correlated well with those held for the longer period. Our method was to set up the cultures as we usually did in the morning, then check them two or three times during the evening shift. We prepared special forms for the evening staff to record the zones after 6, 8, and 10 hours. The plates were not dependable at less than 8 hours, but at 8 to 10 hours, the majority of them became easy to read.

We also realized that a rapid result is no better than a slow one if it leaves the laboratory quickly but is delayed getting to the attending physician. With our routine, culture results reported in the morning often got to the nurses' station after the physician had made rounds. We had to try to get these results onto patient charts before 8 a.m.

A protocol implemented last year (see Figure II) is designed to meet the morning deadline. Amazingly simple, it involved only a slight rearrangement of duties for the evening technologist, who now reads susceptibilities that have been set up during the day and sends out reports for charting by night shift nurses. In addition, the evening technologist checks cultures set up by the day shift and sets up necessary susceptibilities. It is rare for this staff member, who usually works chemistry and hematology, not to have time to help in microbiology. In larger laboratories where microbiology is a separate department, an aide might perform similar duties.

Turnaround time for susceptibility reports has decreased by an average of eight hours. Under our old procedures, we averaged 41 hours between the time a culture was set up and the time susceptibility data were reported. As Figure III shows, we have trimmed eight hours from the process, bringing the average turnaround down to 33 hours. Twenty-two per cent of the susceptibility results are out in 24 hours or less.

Thus physicians can begin or alter antibiotic therapy on the basis of laboratory data eight to 24 hours faster. We anticipate that patients' length of stay in the hospital will decrease as a result. We also believe that rapid susceptibility results may decrease use of expensive broad-spectrum antibiotics. Further monitoring will tell us if these important benefits have materialized.

1. Weinstein, M.P.; Reller, L.B.; Murphy, J.R.; and Lichenstein, K.A. The clinical significance of positive blood cultures. Rev. Infect. Dis. 5(1): 35-53, 1983.

2. Wertman, B.G.; Sostrin, S.V.; Pavlova, Z.; and Lundberg, G.D. Why do physicians order laboratory tests? A study of laboratory test request and use patterns. JAMA 243(20): 2080-2082, 1980.

3. Doern, G.V.; Scott, D.R.; and Rashid, A.L. Clinical impact of rapid antimicrobial susceptibility testing of blood culture isolates. Antimicrob. Agents Chemother. 21: 1023-1024, 1982.

Photo: Figure 1 How reporting influences antibiotic therapy

Photo: Figure II A protocol to decrease turnaround time for susceptibility reports

Photo: Figure III Time from culture setup to susceptibility results
COPYRIGHT 1987 Nelson Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1987 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:column; therapy program based on lab results
Author:Ainsworth, Kristie
Publication:Medical Laboratory Observer
Date:May 1, 1987
Words:1320
Previous Article:How we customized our lab for DRGs.
Next Article:A tailor-made workload recording program.
Topics:


Related Articles
Microcomputers in microbiology: a matter of special needs.
A lab-pharmacy push to cut drug therapy costs.
Why did our reporting trends suddenly change? What a microbiology lab learned about resistance and about a jump in recovery of one pathogen had a...
Rapid microbiology reporting: myth or reality?
Making organism identification cost-effective.
Improve microbiology procedures and utilization - while saving money.
Clinical failures of linezolid and implications for the clinical microbiology laboratory. (Commentary).
Aggregated antibiograms and monitoring of drug-resistant Streptococcus pneumoniae.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters