Morbidity and cost burden of methicillin-resistant Staphylococcus aureus in early onset ventilator-associated pneumonia
© Shorr et al.; licensee BioMed Central Ltd. 2006
Received: 16 February 2006
Accepted: 3 May 2006
Published: 29 June 2006
To gain a better understanding of the clinical and economic outcomes associated with methicillin-resistant Staphylococcus aureus (MRSA) infection in patients with early onset ventilator-associated pneumonia (VAP), we retrospectively analyzed a multihospital US database to identify patients with VAP over a 24 month period (2002–2003).
Data recorded included physiologic, laboratory, culture, and other clinical variables from 59 institutions. VAP was defined as new positive respiratory culture after at least 24 hours of mechanical ventilation (MV) and the presence of primary or secondary ICD-9-CM diagnosis codes of pneumonia. Outcomes measures included in-hospital morbidity and mortality for the population overall and after onset of VAP (duration of MV, intensive care unit [ICU] stay, in-hospital stay, and case mix and severity-adjusted operating cost). The overall cost was calculated at the hospital level using the Center for Medicare and Medicaid Services Cost/Charge Index for each calendar year.
A total of 499 patients were identified as having VAP. S. aureus was the leading organism (31% of isolates). Patients with MRSA were significantly older than patients with methicillin-sensitive Staphylococcus aureus (MSSA; median age 74 versus 67 years, P < 0.05) and more likely to be medical patients. Compared with MSSA patients, MRSA patients on average consumed excess resources of 4.4 (95% confidence interval 0.6–8.2) overall MV days, 3.8 (-0.5 to +8.0) days of inpatient length of stay (LOS), 5.3 (1.0–9.7) ICU days, and US$7731 (-US$8393 to +US$23,856) total cost after controlling for case mix and other factors. Furthermore, MRSA patients needed excess resources after the onset of VAP (4.5 [95% confidence interval 1.0–8.1] MV days, 3.7 [-0.5 to +8.0] inpatient days, and 4.4 [0.4–8.4] ICU days) after controlling for the same case mix and admission severity covariates.
S. aureus remains a common cause of VAP. VAP due to MRSA was associated with increased overall LOS, ICU LOS, and attributable ICU LOS compared with MSSA-related VAP. Although not statistically significant because of small sample size and large variation, the attributable excess costs of MRSA amounted to approximately US$8000 per case after controlling for case mix and severity.
Ventilator-associated pneumonia (VAP) is a major nosocomial infection in the intensive care unit (ICU), affecting between 10% and 20% of patients who receive more than 48 hours of mechanical ventilation (MV) . Several studies [2–5] have documented that VAP significantly increases hospital and ICU length of stay (LOS), duration of MV, and hospital costs. VAP may also increase mortality, although this remains controversial .
Staphylococcus aureus is the leading cause of VAP in Europe and ranks alongside Pseudomonas aeruginosa as the most prevalent cause of VAP in US hospitals [6, 7]. The increasing prevalence of antibiotic-resistant strains is of concern, with more than 60% of S. aureus isolates recovered from the ICU now resistant to methicillin (methicillin-resistant Staphylococcus aureus [MRSA]) . MRSA appears to be associated with worse outcomes than methicillin-sensitive S. aureus (MSSA) infection. In a meta-analysis conducted by Cosgrove and coworkers , MRSA bacteremia was found to increase significantly the risk for death compared with MSSA bacteremia. However, evaluation of the excess burden of MRSA compared with MSSA has also been limited by confounding factors. In one of the largest studies evaluating the impact of MRSA in VAP , the unadjusted mortality rate was significantly higher for MRSA than for MSSA infection. Following adjustment for severity of illness at admission, duration of ICU LOS before VAP diagnosis, and adequacy of antimicrobial treatment, this difference was no longer significant. Combes and coworkers  recently evaluated a group of VAP patients who all received initial appropriate antibiotic therapy, and reported that there was no excess attributable mortality for MRSA versus MSSA.
Although mortality is an important end-point in any analysis of the critically ill, morbidity and costs of critical care are also significant issues . Investigators in the US have reported that MRSA significantly increases LOS and hospital charges compared with MSSA in bacteremia and surgical site infections, with an excess attributable LOS of 2–8 days and excess attributable charges of US$7000–14,000 [13–15]. Morbidity and cost data from analyses of multiple institutional databases specific to MRSA-related pneumonias are more limited and are not US focused.
Given the paucity of local data on the clinical and economic impact of MRSA-related VAP, we conducted a retrospective analysis involving a large US database with two objectives in mind: to identify risk factors associated with the development of VAP among patients admitted to the ICU and to assess the affect of VAP on patient outcomes, including overall and attributable hospital morbidity, mortality, and total costs.
Materials and methods
A retrospective cohort analysis was performed to examine the clinical and economic outcomes associated with MRSA and MSSA in patients with VAP. All available clinical and financial data were obtained for all patients with VAP admitted to 59 US hospitals (16 teaching and 43 nonteaching) between 1 January 2002 and 31 December 2003.
Analyses of the impact of pathogens on MRSA versus MSSA mono-microbial subgroups (total n = 154) were particularly focused on the objective measures of resource utilization, including length of MV, in-patient LOS, and ICU LOS. These dependent variables were analyzed separately as the overall length of event and the length of event from the onset of VAP. Hospital costs for each VAP case were calculated as follows: specific operating cost (defined as the hospital specific cost/charge ratio × hospital total charges); and total cost (defined as [operating cost/charge ratio + capital cost/charge ratio] × total charge). The cost/charge ratios were obtained from the Center for Medicare and Medicaid Services database for each hospital for calendar years 2002 and 2003 . In other words, we computed total costs for each patient by taking the component charges for the hospital stay (for example, pharmacy, laboratory, bed day, and so on) and then multiplying them by the Medicare cost-to-charge ratios. These component costs were then summed to derive the total costs.
Data for the present analysis were obtained from the Cardinal Health Research Database (Cardinal Health, MediQual, Marlborough, MA, USA [formerly MedisGroup]) , a large, multi-institutional database of US acute-care hospitals. Details of this database were reported previously [3, 17–20]. Briefly, the database comprises clinical findings, including patient's clinical history, pathophysiologic findings (such as, vital signs, laboratory test results, culture findings), and physician assessments also includes administrative data imported from hospital information systems [3, 7, 17–20]. The clinical data are generally collected for the first two days of hospitalization for all participating hospitals on all of the patients. For a small group of hospitals (n = 59), laboratory data, including culture data, are collected for the first five days of hospitalization. Hospitals that opted for five day data collections did so voluntarily, primarily for the purposes of internal quality improvement. Because of this limitation, we were only able to identify early onset VAP (mostly within five days of hospitalization).
During the study period, a total of 59 US hospitals (16 teaching and 43 nonteaching) collected clinical and culture data for the first 5 days for all of their patients. Among the 59 hospitals, 42 reported early onset VAP cases with positive culture (within five days of hospitalization). These 42 hospitals were included in the final analysis. Among them, 17% had numbers of beds below 100; 76% had between 101 and 300 beds, and 7% had more than 300. The teaching hospitals accounted for 21% of hospitals. VAP was defined as present in patients who were on a ventilator for at least 24 hours with a first positive bacterial respiratory culture after initiation of MV and with either primary or secondary ICD-9-CM (International Classification of Diseases, Ninth Revision, Clinical Modification) diagnosis codes for pneumonia. For patients who had more than one admission to the ICU for VAP during the same hospitalization, the first ICU admission was used in this analysis. Resource utilization (in-hospital LOS, ICU LOS, and duration of MV) included data for the entire hospital stay.
Univariate and multivariate analyses were conducted for all dependent measures. Nonparametric tests were used for continuous variables and χ2 test for dichotomous variables. Ordinary least square multiple regressions were used to estimate the impact of MRSA on resource utilizations. Covariates included risk for death score (the predicted probability of death based on admission severity), severe trauma, major surgery, number of days from admission to onset of VAP, and MRSA. The risk factors used to generate the death propensity score included demographic variables, coexisting conditions (such as, cancer, cerebrovascular disease, liver disease, renal disease, congestive heart failure, immunosuppressed status), physical examination findings (for example, vital signs, altered mental status), laboratory findings (for example, blood urea nitrogen, glucose), and other clinical findings. In the multivariate analysis we also tested the effect of mortality status to examine whether patients who died in hospital had different resource consumption compared with survivors. In addition, we tested the interaction effect of mortality status and MRSA status to determine whether the MRSA effect varied by mortality status. Finally, we tested hospital's teaching status on all of the outcome measures.
All statistical analyses were conducted using SAS software (version 9.01; SAS Institute Inc., Cary, NC, USA). P < 0.05 was considered statistically significant.
Patient characteristics, pathogen distribution, and risk factors for ventilator-associated pneumonia
Patient demographic and clinical characteristics
MRSA (n = 59)
MSSA (n = 95)
Gram negative (n = 182)
Age (median [IQR]; years)
Chronic renal failure
Congestive heart failure
Median number of comorbidities (IQR)
Acute clinical presentation on admission
Albumin <2.4 g/dl
Arterial pH <7.3
Creatinine >1.5 mg/dl
WBC <4.3 or >19.8 × 103/mm3
BUN >55 mg/dl
BUN 40–55 mg/dl
Respiration >39 breaths/min
ASGd ≥ 3
Clinical and economic outcomes
In patients with VAP, the crude in-hospital mortality rate of patients with S. aureus infection was not statistically significantly different from that due to Gram-negative infection (29% versus 36%). Similarly, there were no significant differences in terms of mortality rates between patients with MRSA and those with MSSA (29% versus 36%).
Nonparametric univariate analysis of MRSA versus MSSA: duration of MV, inpatient LOS, ICU LOS, and total cost
MRSA – all (n = 59)
MSSA – all (n = 95)
MRSA Survivors (n = 42)
MSSA Survivors (n = 61)
MRSA nonsurvivors (n = 17)
MSSA nonsurvivors (n = 34)
Total MV (days)
VAP MV (days)
Total inpatient LOS (days)
VAP inpatient LOS (days)
Total ICU (days)a
VAP ICU (days)a
Total cost (US$)
Multivariate analysis of MRSA versus MSSA: duration of MV, inpatient LOS, ICU LOS, and total cost
MRSA estimate (95% CI)
Total MV (days)
4.4 (0.6 to 8.2)
VAP MV (days)
4.5 (1.0 to 8.1)
Total inpatient LOS (days)
3.8 (-0.5 to +8.0)
VAP inpatient LOS (days)
3.7 (-0.5 to +8.0)
Total ICU (days)a
5.3 (1.0 to 9.7)
VAP ICU (days)a
4.4 (0.4 to 8.4)
Total cost (US$)
7731 (-8393 to +23,856)
The present investigation utilized a large US database to analyze retrospectively the clinical and health economic impact of VAP caused by MRSA versus MSSA infection. We found that MRSA-related VAP was associated with significantly increased overall hospital LOS, ICU LOS, and attributable ICU LOS as compared with MSSA. Although not statistically significant because of small sample size and large variance, the attributable excess cost reached approximately US$8000 per case after controlling for case mix and severity.
The present analysis of objectively measured resource utilization and hospital level costs for the overall and attributable impact of MRSA confirms the findings of two other retrospective inpatient US database analyses [7, 21]. Kollef and coworkers  observed that S. aureus was a major pathogen not only in VAP but also in community-acquired, hospital-acquired, and health care associated pneumonias. In that previous study, pneumonias acquired in the health care and hospital settings were associated with greater LOS and increased cost compared with community-acquired pneumonia, with S. aureus found to be the only pathogen that correlated with mortality after controlling for confounding factors . Noskin and colleagues  reported that S. aureus infection increased the average hospital LOS by threefold, with five times the risk for in-hospital mortality and threefold greater hospital charges compared with inpatients without S. aureus infection. The latter study exclusively relied on ICD-9-CM codes, which did not differentiate between MRSA and MSSA, whereas our study used pathophysiologic data (culture) to confirm the pathogens. Furthermore, the study conducted by Noskin and coworkers used ICD-9-CM codes for severity adjustment, whereas our study used clinical data (vital signs, laboratory, and other clinical findings) for more accurate severity adjustment. Finally, the clinical database we used allowed us to assess the morbidity impact of MRSA after the onset of VAP.
Compared with a recent study using clinical data investigating the impact of MRSA on VAP in France , which showed that MRSA prolonged the median ICU LOS by 11 days compared with MSSA, independent of illness severity, the present study in the US patient population found shorter excess ICU LOS (about four days) after controlling for confounders. This might be due to possible differences in VAP case mix between the two studies and in ICU management between the USA and France.
Measuring the impact of MRSA is crucial for clinicians and administrators aiming to understand the added resource use arising from its diagnosis, prevention, and treatment. Conceivably, even minimally effective preventive strategies could lead to substantial cost savings, and approaches that combine infection prevention and decreasing the prevalence of MRSA have the potential to yield major savings. In the case of MRSA-related VAP, silver coated endotracheal tubes that purportedly prevent VAP or lymphostatin, which is used to prevent patient colonization by MRSA, may find roles in the critical care setting pending successful demonstration of efficacy in clinical trials [23, 24]. Although measures to control the spread of MRSA may seem cumbersome to providers, the findings of this and other studies suggest that any costs associated with the implementation of such measures may quickly be offset by cost savings in the ICU [25, 26]. Any therapeutic intervention directed against nosocomial MRSA infection should be one component of a comprehensive strategy that includes hygiene controls, early pathogen detection, isolation of MRSA-infected patients, and elimination of inappropriate antibiotic use . With respect to the latter, drug acquisition costs are a small component of overall hospital costs and may in fact be trivial compared with the costs associated with prolonged ICU LOS as a result of initial inadequate empiric antibiotic selection.
Our analysis has several limitations. First, we only identified early onset VAP cases within five days of hospitalization. This probably underestimated the prevalence of VAP and the impact of MRSA because many VAP cases are likely to develop later during hospitalization. Second, the retrospective design is associated with a number of limitations, and the present study is as subject to these as any study of this design. An example is that coding practices relating to the use of the ICD-9-CM system vary across hospitals and may be influenced by financial incentives. This introduces uncertainty and limits our findings, despite the fact that we used the concomitant presence of positive respiratory culture and ICD-9-CM codes to define the VAP cases. Third, because we used positive culture to define cases, we were unable to estimate the potential VAP cases in general and MRSA cases in particular. Fourth, given the unique aspects of the hospitals we studied and the other issues noted above, the generalizability of our findings may be limited. Finally, we could not assess the possible confounding factor of appropriate use of antibiotics because this information was not available.
Despite the limitations outlined above, the present study highlights the clinical and economic importance of MRSA in VAP. The data suggest that a diagnosis of MRSA-related VAP is an important determinant of excess hospital and ICU LOS, and attributable costs compared with MSSA infection. These data support the need to develop and implement effective strategies for the diagnosis, treatment, and prevention of MRSA-related VAP in order to reduce patient morbidity and costs associated with provision of health care.
Morbidity and cost data from analyses of large, multi-institutional databases specific to MRSA infection in the USA are limited.
We quantified the clinical and economic outcomes associated with MRSA infection in patients with early onset VAP.
S. aureus was found to be a common cause of VAP. Compared with VAP due to MSSA, VAP due to MRSA was associated with increased overall and attributable resource utilization, prolonged length of stay, and total costs of hospitalization.
= International Classification of Diseases, Ninth Revision, Clinical Modification
= intensive care unit
= length of stay
= methicillin-resistant Staphylococcus aureus
= methicillin-sensitive Staphylococcus aureus
= mechanical ventilation
= ventilator-associated pneumonia.
We thank the following Cardinal Health Clinical Research Group members for their dedicated contributions in database management, analysis, and technical support: Rosemary Kirousis, MPH; Michael Peng, MPH, PhD; Steve Kurtz, MS; Karen Derby, MA; Alisa Goetz, Pharm.D; and Linda Hyde, RHIA.
This study was supported by a research grant from Pfizer Inc.
- Safdar N, Dezfulian C, Collard HR, Saint S: Clinical and economic consequences of ventilator-associated pneumonia: a systematic review. Crit Care Med 2005, 33: 2184-2193. 10.1097/01.CCM.0000181731.53912.D9View ArticlePubMedGoogle Scholar
- Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C: The attributable morbidity and mortality of ventilator-associated pneumonia in the critically ill patient. The Canadian Critical Trials Group. Am J Respir Crit Care Med 1999, 159: 1249-1256.View ArticlePubMedGoogle Scholar
- Rello J, Ollendorf DA, Oster G, Vera-Llonch M, Bellm L, Redman R, Kollef MH: Epidemiology and outcomes of ventilator-associated pneumonia in a large US database. Chest 2002, 122: 2115-2121. 10.1378/chest.122.6.2115View ArticlePubMedGoogle Scholar
- Warren DK, Shukla SJ, Olsen MA, Kollef MH, Hollenbeak CS, Cox MJ, Cohen MM, Fraser VJ: Outcome and attributable cost of ventilator-associated pneumonia among intensive care unit patients in a suburban medical center. Crit Care Med 2003, 31: 1312-1317. 10.1097/01.CCM.0000063087.93157.06View ArticlePubMedGoogle Scholar
- Hugonnet S, Eggimann P, Borst F, Maricot P, Chevrolet JC, Pittet D: Impact of ventilator-associated pneumonia on resource utilization and patient outcome. Infect Control Hosp Epidemiol 2004, 25: 1090-1096. 10.1086/502349View ArticlePubMedGoogle Scholar
- Chastre J, Fagon J-Y: Ventilator-associated pneumonia. Am J Respir Crit Care Med 2002, 165: 867-903.View ArticlePubMedGoogle Scholar
- Kollef MH, Shorr A, Tabak YP, Gupta V, Liu LZ, Johannes RS: Epidemiology and outcomes of healthcare-associated pneumonia. Results from a large US database of culture-positive pneumonia. Chest 2005, 128: 3854-3862. 10.1378/chest.128.6.3854View ArticlePubMedGoogle Scholar
- Fridkin SK, Hill HA, Volkova NV, Edwards JR, Lawton RM, Gaynes RP, McGowan JE: Temporal changes in prevalence of antimicrobial resistance in 23 US hospitals. Emerg Infect Dis 2002, 8: 697-701.PubMed CentralView ArticlePubMedGoogle Scholar
- Cosgrove SE, Sakoulas G, Perencevich EN, Schwaber MJ, Karchmer AW, Carmeli Y: Comparison of mortality associated with methicillin-resistant and methicillin-susceptible Staphylococcus aureus bacteremia: a meta-analysis. Clin Infect Dis 2003, 36: 53-59. 10.1086/345476View ArticlePubMedGoogle Scholar
- Zahar J-R, Clec'h C, Tafflet M, Garrouste-Orgeas M, Jamali S, Mourvillier B, Lassence AD, Descorps-Declere A, Adrie C, Costa de Beauregard M-A, et al.: Is methicillin resistance associated with a worse prognosis in Staphylococcus aureus ventilator-associated pneumonia? Clin Infect Dis 2005, 41: 1224-1231. 10.1086/496923View ArticlePubMedGoogle Scholar
- Combes A, Luyt CE, Fagon JY, Wollf M, Trouillet JL, Gibert C, Chastre J: Impact of methicillin resistance on outcome of Staphylococcus aureus ventilator-associated pneumonia. Am J Respir Crit Care Med 2004, 170: 786-792. 10.1164/rccm.200403-346OCView ArticlePubMedGoogle Scholar
- Ewig S, Torres A: Prevention and management of ventilator-associated pneumonia. Curr Opin Crit Care 2002, 8: 58-69. 10.1097/00075198-200202000-00010View ArticlePubMedGoogle Scholar
- Cosgrove SE, Qi Y, Kaye KS, Harbarth S, Karchmer AW, Carmeli Y: The impact of methicillin resistance in Staphylococcus aureus bacteremia on patient outcomes: mortality, length of stay, and hospital charges. Infect Control Hosp Epidemiol 2005, 26: 166-174. 10.1086/502522View ArticlePubMedGoogle Scholar
- Abramson MA, Sexton DJ: Nosocomial methicillin-resistant and methicillin-susceptible Staphylococcus aureus primary bacteremia: at what costs? Infect Control Hosp Epidemiol 1999, 20: 408-411. 10.1086/501641View ArticlePubMedGoogle Scholar
- Engemann JJ, Carmeli Y, Cosgrove SE, Fowler VG, Bronstein MZ, Trivette SL, Briggs JP, Sexton DJ, Kaye KS: Adverse clinical and economic outcomes attributable to methicillin resistance among patients with Staphylococcus aureus surgical site infection. Clin Infect Dis 2003, 36: 592-598. 10.1086/367653View ArticlePubMedGoogle Scholar
- Acute Inpatient Files for Download [http://www.cms.hhs.gov/AcuteInpatientPPS/FFD/list.asp#TopOfPage]
- Iezzoni LI, Ash AS, Coffman GA, Moskowitz MA: Admission and mid-stay MedisGroups scores as predictors of hospitalization charges. Med Care 1991, 29: 210-220. 10.1097/00005650-199103000-00003View ArticlePubMedGoogle Scholar
- Iezzoni LI, Moskowitz MA: A clinical assessment of MedisGroups. JAMA 1988, 260: 3159-3163. 10.1001/jama.260.21.3159View ArticlePubMedGoogle Scholar
- Silber JH, Rosenbaum PR, Schwartz JS, Ross RN, Williams SV: Evaluation of the complication rate as a measure of quality of care in coronary artery bypass graft surgery. JAMA 1995, 274: 317-323. 10.1001/jama.274.4.317View ArticlePubMedGoogle Scholar
- Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, Coley CM, Marrie TJ, Kapoor WN: A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997, 336: 243-250. 10.1056/NEJM199701233360402View ArticlePubMedGoogle Scholar
- Noskin GA, Rubin RJ, Schentag JJ, Kluytmans J, Hedblom EC, Smulders M, Lapetina E, Gemmen E: The burden of Staphylococcus aureus infections on hospitals in the United States. An analysis of the 2000 and 2001 Nationwide Inpatient Sample Database. Arch Intern Med 2005, 165: 1756-1761. 10.1001/archinte.165.15.1756View ArticlePubMedGoogle Scholar
- Shorr AF, Combes A, Kollef MH, Chastre J: Methicillin resistant Staphylococcus aureus prolongs intensive care unit length of stay in ventilator-associated pneumonia despite initially appropriate antibiotic therapy. Crit Care Med 2006, 34: 700-706. 10.1097/01.CCM.0000201885.57697.21View ArticlePubMedGoogle Scholar
- Olson ME, Harmon BG, Kollef MH: Silver-coated endotracheal tubes associated with reduced bacterial burden in the lungs of mechanically ventilated dogs. Chest 2002, 121: 863-870. 10.1378/chest.121.3.863View ArticlePubMedGoogle Scholar
- Kokai-Kun JF, Walsh SM, Chanturiya T, Mond JJ: Lysostaphin cream eradicates Staphylococcus aureus nasal colonization in a cotton rat model. Antimicrob Agents Chemother 2003, 47: 1589-1597. 10.1128/AAC.47.5.1589-1597.2003PubMed CentralView ArticlePubMedGoogle Scholar
- Muto CA, Jernigan JA, Ostrowsky BE, Richet HM, Jarvis WR, Boyce JM, Farr M: SHEA guideline for preventing nosocomial transmission of multidrug-resistant strains of Staphylococcus aureus and enterococcus. Infect Control Hosp Epidemiol 2003, 24: 362-386. 10.1086/502213View ArticlePubMedGoogle Scholar
- Puzniak LA, Leet T, Mayfield J, Kollef MH, Mundy LM: To gown or not to gown: the effect on acquisition of vancomycin resistant enterococci. Clin Infect Dis 2002, 35: 18-25. 10.1086/340739View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.