HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (21) Citing Articles via Google Scholar Google Scholar Articles by Silber, S. H. Articles by Okafo, T. Search for Related Content PubMed PubMed Citation Articles by Silber, S. H. Articles by Okafo, T.

Early Administration of Antibiotics Does Not Shorten Time to Clinical Stability in Patients With Moderate-to-Severe Community-Acquired Pneumonia^*

^* From the Department of Emergency Medicine, New York Methodist Hospital, Brooklyn, NY.

Correspondence to: Steven H. Silber, DO, MS, Department of Emergency Medicine, New York Methodist Hospital, 506 Sixth St, Brooklyn, NY 11215; e-mail: sts9005{at}nyp.org

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Study objective: To determine if there is a statistically significant difference in the time to clinical stability (TCS) between those patients with moderate-to-severe (MTS) community-acquired pneumonia (CAP) who received their antibiotics within 4 h and those who received their antibiotics after 4 h.

Design: Prospective observational study.

Setting: A large metropolitan teaching institution with 62,000 annual emergency department visits from May 1999 through January 2001.

Patients: Patients were 21 year with MTS CAP as defined by the Pneumonia Patient Outcomes Research Team (PORT).

Interventions: Triage-to-needle time (group 1, 0 to 240 min; group 2, 241 to 480 min; and group 3, > 480 min) was the independent variable, and TCS was the dependent variable. Our hypothesis was that door-to-needle time < 4 h would result in TCS reduction of 0.5 days.

Measurements: Statistical analysis was performed using the two-tailed Student t test, analysis of variance, and multiple linear regression; p < 0.05 was considered significant.

Results: Four hundred nine patients with MTS CAP achieved clinical stability during their hospital stay. Fifty-four percent of patients received antibiotics within 4 h. The mean time to receiving antibiotics was 131.46 min (2.19 h) in group 1, 335.52 min (5.59 h) in group 2, and 783.98 min (13.07 h) in group 3. Mean TCS was 3.19 days in group 1, 3.16 days in group 2, and 3.29 days in group 3. There were no statistically significant differences in TCS between the study groups.

Conclusion: The administration of antibiotics within 4 h does not reduce the TCS in adult patients with MTS-CAP, as defined by the PORT group. Future studies using other physiologic parameters should be explored.

Key Words: antibiotics • community-acquired pneumonia • hospitalization guidelines • length of stay • outcomes • pneumonia severity index • time to clinical stability

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Community-acquired pneumonia (CAP) is among the most frequent causes of hospitalization in the United States, and is the sixth-leading cause of death overall. Pneumonia is a leading cause of hospitalizations among the elderly.¹ The annual cost of treating patients > 65 years old for CAP is approximately 4.8 billion dollars; the majority of costs stemming from inpatient charges.² Studies on outcome improvement on patients with CAP have focused on time of antibiotic administration as a key in reducing mortality and length of stay (LOS). Antibiotic administration in < 8 h has been shown to reduce 30-day mortality in patients > 65 years old.³ However, other outcomes, such as in-hospital mortality, time to clinical stability (TCS), and LOS, have not been conclusively linked to the timing of antibiotic administration. Despite the lack of evidence correlating antibiotic timing with these outcome measures, the Center for Medicare and Medicaid Services (CMS) has proposed a 4-h metric of antibiotic administration as a quality measure in the treatment of CAP.⁴

The relationship between the timing of antibiotic administration and LOS has been limited by factors such as physician practice style and social stays in the hospital.⁵ TCS was defined in a large prospective observational study⁶ allowing a more objective measure of patient outcome.

The objective of this study was to determine if there is a statistically significant difference in TCS between those patients with moderate-to-severe (MTS) CAP who received their antibiotics within 4 h and those who received their antibiotics after 4 h. Using the 4-h antibiotic metric should provide useful data in determining the significance of a CMS change in its quality indicators.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
This was a prospective study of patients 21 years old admitted to the hospital via the emergency department with MTS CAP, as defined by the Pneumonia Patient Outcomes Research Team (PORT) classification system,⁷ between May 1999 and January 2001. The study was performed at a large, metropolitan teaching institution with 62,000 annual emergency department visits. This project was approved by the hospital Institutional Review Board.

All patients 21 years old admitted through the emergency department with a diagnosis of pneumonia were eligible for the study. Patients < 21 years old were not included because, as per hospital policy, these patients are seen in a separate pediatric emergency department.

Patients were excluded from review if they did not meet the criteria for CAP, as defined by Fine et al.⁷ This included hospitalization within 1 week prior to hospital admission, hematologic malignancy, immunosuppressive illness, or initial chest radiographic findings not consistent with pneumonia. Immunosuppressive illness was defined as HIV infection, use of > 10 mg/d of prednisone or other immunosuppressive agents, active treatment for cancer, history of organ transplantation, active tuberculosis, or cystic fibrosis.

The hospital utilizes a clinical policy for the administration of antibiotics for CAP consistent with the recommendations of the American Thoracic Society⁸ and the Infectious Disease Society of America.⁹ Patients were also excluded from the study if they received IV antibiotics before hospital presentation or received antibiotics not consistent with American Thoracic Society or Infectious Disease Society of America guidelines.

Patients were identified at time of hospital admission and were followed up through discharge or death. Data were collected for demographic information, triage time, time to antibiotics, TCS, LOS, initial choice of antibiotics, and in-hospital outcome.

PORT classifications were calculated using a pneumonia severity index (PSI) derived from a prediction rule by Fine et al⁷ based on analysis of 14,199 adult inpatients with CAP. The PSI was used to categorize patients into PORT classifications 1 to 5, with classes 3 to 5 defined as having MTS CAP.

Time to antibiotics was defined as the time from triage to the initiation of the initial dose of antibiotics. TCS was calculated using the first day six clinical parameters were met by the patient. The parameters, studied by Halm et al⁶ in 686 patients, are listed in Table 1 . The parameters were required to have been maintained for a full 24 h by the patient in order to meet the definition of stability.

To account for this difference, day 1 was divided into quartiles. Patients triaged between midnight and 6:00 AM were considered to have spent 1 complete day (1.00) in the hospital on day 1. Patients triaged from 6:00 AM to noon, from noon to 6:00 PM, and from 6 PM to midnight were considered to have spent 0.75 days, 0.5 days, and 0.25 days in the hospital on day 1, respectively.

Our primary study group was defined as those patients having MTS CAP discharged alive and having achieved clinical stability during their hospitalization. Study groups were defined based on the timing of administration of their first dose of antibiotics. Group 1 was defined as those patients receiving antibiotics within 4 h (1 to 240 min) of triage, group 2 as those receiving antibiotics from 4 to 8 h (241 to 480 min) after triage, and group 3 as those receiving antibiotics > 8 h ( 481 min) from triage.

The primary study outcome was the comparison between the study groups for differences in the mean TCS. Secondary study outcomes compared study groups for differences in mean LOS, mean PSI, and mortality. Study groups were also categorized into subgroups by PORT class, and compared for differences.

The subgroup of patients that died (mortality group) was similarly analyzed for mean PSI using the 4-h metric. Mean PSI of the mortality group was compared to the mean PSI of our primary study group. Patients who did not achieve clinical stability were analyzed separately for timing of antibiotic administration, using the same 4-h metric as the independent variable.

Two-group comparisons of means were done using the two-tailed Student t test. Three-group comparisons of means were done using one-way randomized ANOVA models. Following significant ANOVA, the Tukey HSD test was used to determine the significance of pairwise differences between means. When treating time to treatment as a continuous variable in predicting TCS and LOS, multiple linear regression models were used.

We calculated that 199 patients per group would be required to attain an 80% power to detect the difference of 0.5 days of the geometric mean of TCS between those groups receiving antibiotics within 4 h and those receiving antibiotics > 4 h; p < 0.05 was significant. All analytical procedures were performed using the STATA 8 statistical package (Stata Corporation; College Station, TX).

	Results

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Six hundred three patients were entered into the study. Seventy-five patients (13%) were eliminated due to administration of antibiotics outside the hospital guidelines. One hundred eighteen patients (20%) failed to achieve clinical stability during their hospital stay, including 68 mortalities (11%) in patients with MTS CAP. Four hundred ten patients (68%) achieved clinical stability and fell in the MTS group; this became our primary study group (Table 2 ).

Descriptive statistics of the study groups are listed in Table 3 . The mean age was 78.4 years (range, 33 to 101 years); there were 265 female and 144 male patients in the MTS study sample. One hundred forty-four patients were from skilled nursing facilities.

ANOVA was performed to analyze mean TCS between the three study groups. No statistically significant differences were found (p = 0.984). Using a linear regression model with the outcome TCS and the predictor "time to treatment," there was no statistically significant linear relationship found between the two variables (95% confidence interval [CI] for slope, - 0.002 to 0.002).

Group 1 was compared to group 2 for mean TCS, utilizing the Student t test. The difference was not statistically significant (p = 0.982; 95% CI, - 0.8562704 to 0.8370962).

Group 1 was then compared to groups 2 plus 3 to compare mean TCS between those patients receiving antibiotics within 4 h and all patients receiving antibiotics > 4 h. There were no significant differences (p = 0.923; 95% CI, - 0.8562704 to 0.8370962).

The mean LOS was 9.01 days (SD, 6.00 days) in group 1, 10.21 days (SD, 7.18 days) in group 2, and 9.75 days (SD, 6.41 days) in group 3. There were no statistically significant differences between the three study groups (p = 0.224). Using a linear regression model with the outcome LOS and the predictor time to treatment, there was no statistically significant linear relationship between the variables (95% CI for slope, - 0.0007 to 0.005).

Group 2 trended to have a longer mean LOS than group 1, but this was not found to be statistically significant (p = 0.0891; 95% CI, - 2.58839 to 1844418). There was also no significant difference in mean LOS between group 1 and groups 2 plus 3 (p = 0.0934; 95% CI, - 2.3388 to 0.1822).

The overall mean PSI of the primary study group was 116.26 (SD, 32.98). The mean PSI was 116.21 (SD, 33.56) in group 1, 115.98 (SD, 32.15) in group 2, and 117.39 (SD, 33.25) in group 3. There were no significant differences between study groups (p = 0.96).

Table 5 reflects the subgroup analysis of study groups by PORT class. When analyzed for mean TCS, mean LOS, and mean PSI, no significant differences were found.

The 50 patients with MTS CAP who failed to meet clinical stability were categorized into two additional groups: group 4 (antibiotics received within 4 h), and group 5 (antibiotics received after 4 h). These groups were analyzed for LOS. There were 25 patients in each group. The LOS for group 4 was 14.5 days, and group 5 was 13.6 days. The difference in LOS for these patients was not statistically significant.

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Antibiotics are the primary weapons of physicians to improve to the clinical outcome of CAP. Utilizing them as early as possible makes good clinical sense, but in the end might not be the primary factor in the recovery time of the patient. Other factors involved in the achievement of clinical stability are nutritional status, comorbid conditions, the strength of the immune system, and the time from initial onset of illness to the time of hospital presentation. The timing of antibiotic administration is one indicator of patient care, but other interventions may also play significant roles, such as adequate caloric intake and fluid replacement.

Our study suggests that earlier antibiotic administration did not shorten hospital LOS or allow earlier achievement of clinical stability. However, this does not imply that antibiotics should not be administered expeditiously in caring for patients with CAP. Antibiotics administered within 8 h have been shown to reduce 30-day mortality in patients 65 years old, but 30-day mortality and TCS are two very different outcomes. The factors involved in preventing mortality and reducing recovery time in CAP, while both involving quality-of-care issues, are not the same.

Improving the quality of care and reducing cost of CAP treatment have been visited by several investigators. Studies^{3 5 10 11} have shown that hospital admission rates can be decreased by utilization of a CAP clinical pathway, but did not have definitive results on which factors affect inpatient stay. Inpatient stay accounts for the majority of the 4.8 billion dollars that is cited for treatment of CAP. Naturally, there is focus on interventions that reduce the length of hospitalization and hence reduce costs. Variances of hospital LOS due to physician practices, hospital costs, managed-care influences, and patient social factors make an association between LOS and timing of antibiotic administration difficult to subject to scientific studies.^{5 12} Recently, Battleman et al,¹³ showed that patients administered antibiotics in the emergency department (time to antibiotics, 3.5 ± 1.4 h) have a significantly shorter LOS than those patients administered antibiotics on the inpatient unit (time to antibiotics, 9.5 ± 3.0 h). Door-to-needle time was associated with prolonged LOS based on a multivariate analysis using 8-h intervals (multivariate odds ratio, 1.75 per 8 h).

The implementation of CAP clinical pathways has been studied as a way to reduce the time to antibiotic administration to < 8 h. Marrie et al¹⁰ demonstrated a reduction in hospital stay after the implementation of a hospital-based clinical pathway, but the reduction in hospital stay for all diagnoses in the same time period and the absence of a control group prevented a clear association to reduced LOS as a result of the pathway.

The Center for Medicare and Medicaid Services (CMS), formerly Health Care Finance Administration, has taken great interest in the process-outcome relationship of the timing of antibiotic administration on CAP. Current CMS quality indicators mandate the administration of antibiotics within 8 h.¹⁴ This allows physicians time to perform an adequate workup and clear up diagnostic uncertainties. Their proposed seventh scope of work of their national pneumonia quality improvement changes the quality indicator from 8 to 4 h.⁴

At this time, there does not appear to be sufficient evidence in the literature to justify the change. Neither the study by Meehan et al³ nor the study by Battleman et al,¹³ both cited as references for the change, have demonstrated a significant impact of antibiotics initiated earlier than 8 h on 30-day mortality or hospital LOS.

A clear relation between the timing of antibiotic administration and improvement in the process of care of all patients with CAP needs to be shown before unreasonable mandates for patient care are initiated and hospitals are penalized for not being in compliance. This would be the case if the antibiotic indicator extends below the 8-h metric and does not take into account pneumonia severity. The focus of quality improvement must remain on interventions that actually improve the quality of patient care, not on unsubstantiated random time measurements of interventions.

The importance of clearing diagnostic uncertainty and allowing timely delay of antibiotic administration has the advantages of improving diagnostic accuracy.^{15 16} Also, eliminating the pressure of administering antibiotics prior to completion of a reasonable workup might allow physicians to treat viral CAP or influenza without antibiotics. Certainly with the pressure of administering antibiotics as quickly as possible, patients with negative chest radiograph findings and symptoms of pneumonia are unnecessarily receiving antibiotics in the emergency department. Eliminating unnecessary antibiotic use is a measure that would certainly reduce hospital expenditure and antibiotic resistance.

Our study does have limitations. We chose to use the PORT classification to define our study groups. However, this classification was originally designed to predict the probability of 30-day mortality in patients > 65 years old. Extrapolation to match inpatient clinical status may not be accurate. Time-to-treatment group comparisons and linear regression analysis might be more accurately compared using more proven physiologic indicators, such as acute physiology and chronic health evaluation.¹⁷

Evaluation of other variables might also assist in matching clinical status. For example, duration of symptoms prior to hospital presentation was not accounted for in the evaluation of time to treatment. One patient might have had pneumonia for 3 days prior to arrival and received antibiotics within the 4-h metric, while a second patient might have had pneumonia for 1 day prior to arrival and received antibiotics after the 4-h metric. It would appear in our study that the first patient received antibiotics sooner; actually, this patient received antibiotics 2 days later.

We did not review data on patients receiving antibiotics inconsistent with hospital guidelines, nor did we separately evaluate those patients treated with oral antibiotics prior to arrival. These factors may have introduced bias into the study.

Our study found no differences in the mean PSI of PORT classes between the three study groups and a significant difference between mean PSI of our primary study group and that of the mortality group. Although interesting, PSI has not been validated as an acute physiologic instrument. It would be interesting to compare acute physiology and chronic health evaluation scores to PSI and correlate them with TCS and LOS. This would assist in the determination of the actual implications of the PSI distribution.

Clinical outcome, as measured by TCS, is a model for the optimal LOS based on limited adverse outcomes, but it is not the criterion standard for patient discharge. As an objective indicator of patient status, it suited our needs. Other clinical outcome variables might be more sensitive to the timing of antibiotic administration. Studies on the functional recovery of pneumonia might be a more sensitive indicator for the safety of patient discharge and might have altered our results to favor a longer LOS for patients receiving antibiotics later in their hospital stay.¹⁸

Another objective indicator of patient status is mortality. Our study was unable to assess the impact on the timing of antibiotic administration on inpatient mortality. Mortality in our study was within the predicted range by PORT scores. There was a trend for greater mortality in the group that received antibiotics in < 4 h, but this included many patients that died within 24 h of hospital presentation.

Looking for factors that improve the quality of care, reduce costs, and improve efficiency of patient care is a worthwhile endeavor. In CAP, antibiotic administration within 8 h is a reasonable expectation, given that it has been shown to reduce 30-day mortality in patients > 65 years old. However, since our study did not prove that earlier administration reduced TCS, clinicians should not be held to a door-to-needle time of < 8 h.

	Conclusion

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 
Antibiotics administered within 4 h do not reduce the TCS in adult patients with MTS CAP, as defined by the PORT classification system. Future prospective studies using multivariate analysis of indicators such as duration of symptomatology, comorbid conditions, and acute inpatient physiologic profiles should be performed to validate the findings of this study.

	Footnotes

 
Abbreviations: ANOVA = analysis of variance; CAP = community-acquired pneumonia; CI = confidence interval; CMS = Center for Medicare and Medicaid Services; LOS = length of stay; MTS = moderate to severe; PORT = Pneumonia Patient Outcomes Research Team; PSI = pneumonia severity index; TCS = time to clinical stability

Received for publication October 3, 2002. Accepted for publication May 30, 2003.

	References

TOP Abstract Introduction Materials and Methods Results Discussion Conclusion References

 

1. Bartlett, JG, Breiman, R, Mandell, LA, et al (1998) Community-acquired pneumonia in adults: guidelines for management. Clin Infect Dis 26,811-838[ISI][Medline]
2. Niederman, MS, McCombs, JS, Unger, AN, et al The cost of treating community-acquired pneumonia. Clin Ther 1998;20,820-837[CrossRef][ISI][Medline]
3. Meehan, TP, Fine, MJ, Krumholtz, HM, et al Quality of care, process, and outcomes in elderly patients with pneumonia. JAMA 1997;23,2080-2084
4. National Pneumonia Quality Improvement Project. Oklahoma Foundation for Medical Quality. Available at: http://www.nationalpneumonia.org/perf-meas-7SOW/perf-meas-7sow.html. Accessed July 22, 2002
5. Fine, MJ, Singer, DE, Phelps, AL, et al Differences in length of stay in patients with community-acquired pneumonia: a prospective four hospital study. Med Care 1993;31,371-380[CrossRef][ISI][Medline]
6. Halm, EA, Fine, MJ, Marrie, TJ, et al Time to clinical stability in patients hospitalized with community-acquired pneumonia. JAMA 1998;279,1452-1457[Abstract/Free Full Text]
7. Fine, MJ, Auble, TE, Yealy, DM, et al A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997;336,243-250[Abstract/Free Full Text]
8. Niederman, MS, Mandell, LA, Anzueto, A, et al American Thoracic Society. Guidelines for the management of adults with community-acquired pneumonia. Am J Respir Crit Care Med 2001;163,1730-1754[Free Full Text]
9. Bartlett, JG, Scott, DF, Mandell, LA, et al IDSA practice guidelines for the management of community-acquired pneumonia in adults. Clin Infect Dis 2000;31,347-382[CrossRef][Medline]
10. Marrie, TJ, Lau, CY, Wheeler, SL, et al A controlled trial of a critical pathway for treatment of community-acquired pneumonia. JAMA 2000;283,749-755[Abstract/Free Full Text]
11. Benenson, R, Magalski, A, Cavanaugh, S, et al Effect of a pneumonia clinical pathway on time to antibiotic treatment, length of stay, and mortality. Acad Emerg Med 1999;6,1243-1248[ISI][Medline]
12. Fine, MJ, Medser, AR, Stone, RA, et al The hospital discharge decision for patients with community-acquired pneumonia. Arch Intern Med 1997;157,47-56[Abstract]
13. Battleman, DS, Callahan, M, Thaler, HT Rapid antibiotic delivery and appropriate antibiotic selection reduce length of stay of patients with community acquired pneumonia. Arch Intern Med 2002;162,682-688[Abstract/Free Full Text]
14. Health Care Quality Improvement Program. Medicare Priorities, Version 5.2; page 33. Health Care Financing Administration, Baltimore, MD: U. S. Department of Health and Human Services, 2000. Available at: http://www.masspro.org/healthpro/hcqip/priority.pdf. Accessed October 10, 2003
15. Brown, PD, Lerner, SA Community-acquired pneumonia. Lancet 1998;352,1295-1302[CrossRef][ISI][Medline]
16. Niederman, MS, Bass, JB, Jr, Campbell, GD, et al Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy; American Thoracic Society. Am Rev Respir Dis 1993;148,1418-1426[ISI][Medline]
17. Knaus, WA, Wagner, DP, Draper, EA, et al The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991;100,1619-1636[Abstract/Free Full Text]
18. Metlay, JP, Fine, MJ, Schultz, R, et al Measuring symptomatic and functional recovery in patients with community-acquired pneumonia. J Gen Intern Med 1997;12,423-430[CrossRef][ISI][Medline]

This article has been cited by other articles:

				T. J. Marrie and L. Wu Factors Influencing In-hospital Mortality in Community-Acquired Pneumonia: A Prospective Study of Patients Not Initially Admitted to the ICU Chest, April 1, 2005; 127(4): 1260 - 1270. [Abstract] [Full Text] [PDF]

				P. M. Houck, D. W. Bratzler, W. Nsa, A. Ma, and J. G. Bartlett Antibiotic Administration in Community-Acquired Pneumonia Chest, July 1, 2004; 126(1): 320 - 321. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (21) Citing Articles via Google Scholar Google Scholar Articles by Silber, S. H. Articles by Okafo, T. Search for Related Content PubMed PubMed Citation Articles by Silber, S. H. Articles by Okafo, T.