Docstoc

Sepsis Feature Articles Alcohol dependence

Document Sample
Sepsis Feature Articles Alcohol dependence Powered By Docstoc
					Feature Articles

Alcohol dependence is independently associated with sepsis, septic shock, and hospital mortality among adult intensive care unit patients*
James M. O’Brien Jr, MD, MSc; Bo Lu, PhD; Naeem A. Ali, MD; Greg S. Martin, MD, MSc; Scott K. Aberegg, MD, MPH; Clay B. Marsh, MD; Stanley Lemeshow, PhD; Ivor S. Douglas, MD

Objective: To determine the association between alcohol dependence (alcoholism not in remission and/or alcohol withdrawal) and sepsis, septic shock, and hospital mortality among intensive care unit (ICU) patients. Design: Retrospective cohort study. Setting: Two ICUs in an urban hospital. Patients: Patients included 11,651 adult admissions to Denver Health Medical Center from January 1, 1999, to December 31, 2004, with >1 ICU day. Interventions: None. Measurements and Main Results: Of first admissions appearing in the data set (n 9,981), 1,222 (12.2%) had a diagnosis consistent with alcohol dependence. These patients had higher rates of sepsis (12.9% vs. 7.6%, p < .001), organ failure (67.3% vs. 45.8%, p < .001), septic shock (3.6% vs. 2.1%, p .001), and hospital mortality (9.4% vs. 7.5%, p .022) on unadjusted analyses. Patients with alcohol dependence also had fewer hospitalfree days. After adjustment for factors with known association with sepsis, alcohol dependence was associated with sepsis. This association was modified if the patient received (adjusted odds

ratio, 0.92; 95% confidence interval, 0.65–1.31) or did not receive (adjusted odds ratio, 1.91; 95% confidence interval, 1.49 –2.44) red cell transfusions. A general predisposition to infections mediated some, but not all, of this association. Results were similar when repeat admissions were included in the analysis. Alcohol dependence was also associated with septic shock and hospital mortality in multivariable analyses. Among those with liver disease and sepsis, alcohol dependence was associated with more than two-fold increased risk-adjusted odds of hospital mortality (adjusted odds ration, 2.31; 95% confidence interval, 1.26 – 4.24). Similarly, sepsis and liver disease carried higher odds of death for alcohol-dependent patients than for those without alcohol dependence. Conclusions: Alcohol dependence is independently associated with sepsis, septic shock, and hospital mortality among ICU patients. The underlying mechanisms of this association require exploration, as an increased rate of infections mediated some, but not all, of this association. (Crit Care Med 2007; 35:345–350) KEY WORDS: alcohol withdrawal delirium; alcoholism; alcoholrelated disorders; sepsis; shock; septic; critical care

A

lcohol use disorders are associated with many chronic health problems (1), approximately 85,000 deaths per year (2), and an annual cost of $184.6 billion (3). Unhealthy alcohol use covers a spectrum from risky behavior to alcohol dependence with prevalence of 9 –39% among ICU admissions (4, 5). In addition to increasing the risk of critical illnesses, such

as trauma, hypothermia, and pancreatitis, alcohol can lead to respiratory failure from acute alcohol intoxication and alcohol withdrawal syndrome (1). Sepsis is the tenth leading cause of death in the United States (6) with mortality rates exceeding 30% (7). Extremes of age, male gender, and nonwhite race all carry higher rates of sepsis (8). Comorbidities, including liver disease (9), HIV (10),

*See also p. 627. From The Ohio State University Medical Center, Division of Pulmonary, Allergy, Critical Care and Sleep Medicine (JMO, NAA, SKA, CBM); The Ohio State University School of Public Health, Division of Biostatistics (BL); Emory University School of Medicine, Division of Pulmonary, Allergy and Critical Care Medicine (GSM); School of Public Health, Center for Biostatistics, The Ohio State University (SL); and Denver Health Medical Center, University of Colorado Health Sciences Center, Division of Pulmonary Sciences and Critical Care Medicine (ISD). The study was performed at The Ohio State University Medical Center and at Denver Health Medical

Center in concert with The Ohio State University School of Public Health and Center for Biostatistics. Supported, in part, by a Career Development Award (HL075076) from the NHLBI and a Davis/Bremer Award from The Ohio State University College of Medicine (JMO); HL63800 and HL70294 (CBM); and HL070940 (ISD). The authors have not disclosed any potential conflicts of interest. Copyright © 2007 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins DOI: 10.1097/01.CCM.0000254340.91644.B2

and cancer (11), and therapeutic interventions, such as in-dwelling catheters (12) and red cell transfusions (13), have been associated with increased sepsis rates. Alcohol has widespread effects on the immune system (14) and leaves abusers at increased risk of a variety of infections (15, 16). Some studies suggest increased severity in those with established pneumonia (17, 18) and a higher risk of acute lung injury and the acute respiratory distress syndrome in chronic alcohol abusers with other risk factors (19 –21). However, data examining the association between alcohol use disorders and sepsis from nonpulmonary sources are limited. Important alcohol-related comorbidities, such as cirrhosis (9), may be the source of the apparent association. An increased predisposition to infection among patients with alcohol use problems may also mediate an association with sepsis. To address these issues, we conducted a retrospective cohort study
345

Crit Care Med 2007 Vol. 35, No. 2

of intensive care unit (ICU) patients to determine the independent association between alcohol dependence and sepsis, septic shock, and death.

MATERIALS AND METHODS
Study Sample and Subjects. This study used data from the University HealthSystem Consortium (UHC, www.uhc.edu) database, an alliance of academic medical centers designed to allow institutions to benchmark performance against other members. The UHC Clinical Data Base (CDP) includes demographic data, discharge diagnoses, and discharge disposition. Diagnoses and procedures are coded based on the International Classification of Diseases, 9th Revision, Clinical Modifications (ICD-9) and Current Procedural Terminology, respectively. These were abstracted by medical coders at Denver Health Medical Center (DHMC) for billing and reported to UHC. Inclusion criteria included admission to DHMC between January 1, 1999, and December 31, 2004; 1 day spent in an ICU (surgical or medical); and age 18 yrs. Exclusion criteria included admissions transferred to DHMC from another acute care hospital or from an unknown source and admissions with an unknown discharge status. DHMC is a 349-bed academic medical center and “safety net” hospital designated to serve the needs of special populations, including the indigent, chronically mentally ill, and persons addicted to alcohol or other substances. The medical and surgical ICUs are organized in a “closed” management structure with dedicated teams caring for all critically ill patients. Alcohol Dependence, Sepsis, and Septic Shock. Alcohol dependence was identified based on ICD-9 codes in the discharge record. These included codes consistent with ongoing alcoholism and alcohol withdrawal. The diagnosis of sepsis and septic shock was based on previously validated combinations of ICD-9 codes found in the discharge record (8). Septic shock was defined as the presence of a code for sepsis and a code for shock. The Appendix provides details on diagnostic codes. Other organ dysfunction was defined as the presence of previously validated ICD-9 and/or Current Procedural Terminology codes (8). Statistical Analyses. The presence of sepsis was the primary outcome. Diagnoses, hospital mortality, and hospital lengths of stay were extracted from the UHC database. Hospitalfree days were calculated as the number of days alive and not in the hospital within 28 days after admission. Although we ultimately explored the associations of interest in the entire cohort, including repeated admissions, we focused our analyses on admission records from patients appearing for the first time in the data set. We compared demographics and clinical characteristics between those with and without alcohol dependence and sepsis. Secondary outcomes, including septic shock, hos-

pital mortality, length of stay, and discharge location, were examined. In unadjusted analyses, we used Student’s t-test and the KruskalWallis test for continuous variables and Pearson’s chi-square test or Fisher’s exact test for categorical variables, as appropriate. To adjust for the risk for sepsis, we first estimated the unadjusted relationship between alcohol dependence and sepsis. We then adjusted for demographic data and covariates previously described to have an association with sepsis. These included age, race (white vs. nonwhite), gender, cancer (with solid tumors and leukemia/lymphoma considered separately), HIV infection, liver disease, transfusion with red blood cells, and infection or inflammation due to an in-dwelling device. Codes were based upon the Charlson-Deyo system, when available (22) (the Appendix has details). Age was linear in the logit and was included as a continuous variable (data not shown). All of these covariates were retained in the risk-adjusting multivariable logistic regression model, regardless of statistical significance. Additional covariates (diabetes, trauma and burn diagnoses, and admission source) were retained if they altered the adjusted coefficient for alcohol dependence by 10%. To determine whether the association between alcohol dependence and sepsis was due to an increased risk of infection in general, we included a variable for nonseptic, non-HIV infections (ICD-9 codes 000 –139, excluding those included in the sepsis or HIV codes). Once a risk-adjusting model was estimated with parent terms, all second-order interaction terms with alcohol dependence were considered. Interactions were retained in the final risk-adjusting model if the likelihood ratio test produced a p value .01. We refitted the riskadjusting model using a similar strategy for analyses using the secondary outcomes of septic shock. In the analysis of hospital mortality, we were most interested in the association be-

tween alcohol dependence and death in patients with sepsis. Therefore, our primary explanatory variable for these analyses was the interaction between alcohol dependence and sepsis. For hospital mortality, we began with the base model for sepsis and removed covariates that did not significantly improve the likelihood ratio test (p .05). We then considered number of organ failures for inclusion in the model as well as the covariates evaluated for the models for sepsis and septic shock. Our initial risk-adjusting models included only the first admission record for each patient. We refitted the logistic regression models with all admission records using a generalized estimating equation approach. Using the generalized estimating equation, both exchangeable and autoregressive working correlation matrices were used in the analyses, and the results were similar (data not shown). We completed all analyses using SAS for Windows (version 9.1, SAS Institute, Cary, NC). Human Subjects Protection. UHC provided data without identifying information and had no role in the preparation or approval of this manuscript. The Colorado Multiple Institution Review Board approved the conduct of the study without additional informed consent.

RESULTS
The initial cohort included 11,859 adult admissions to DHMC with at least one ICU day. However, 208 (1.8%) hospitalizations involved a transfer to DHMC or had an unknown discharge status and were excluded. Of the 11,651 admission records included in the study, 9,981 (85.7%) represented initial patient admissions, whereas the rest were repeat admissions (Fig. 1). Of first-admissions, 822 (8.2%) ICU admissions had a discharge diagnosis of sepsis. Among the septic patients, 714 (86.9%)

Figure 1. Included and excluded subjects in study cohort. DHMC, Denver Health Medical Center; ICU, intensive care unit.

346

Crit Care Med 2007 Vol. 35, No. 2

Table 1. Differences in demographics and other diagnoses between intensive care unit patients with and without sepsis Sepsis (n 822) Demographics Age, mean (SD) Male gender, n (%) Nonwhite race, n (%) Emergency department admission, n (%) Other diagnoses, n (%) Liver disease HIV Trauma or burn Diabetes Infection or inflammation due to an indwelling device Nonseptic, non-HIV infection Cancer Leukemia or lymphoma Solid tumor Red cell transfusion No Sepsis (n 9159)

p Value

49.6 (16.2) 541 (65.8) 483 (58.8) 642 (78.1) 117 (14.2) 40 (4.9) 274 (33.3) 134 (16.3) 24 (2.9) 401 (48.8) 81 (9.8) 28 (3.4) 56 (6.8) 350 (42.6)

47.0 (17.2) 6061 (66.2) 4939 (53.9) 7031 (76.8) 483 (5.3) 141 (1.5) 4088 (44.6) 1497 (16.3) 38 (0.4) 1410 (15.4) 523 (5.7) 60 (0.7) 466 (5.1) 1774 (19.4)

.015 .83 .008 .38 .001 .001 .001 .97 .001 .001 .001 .001 .033 .001

HIV, human immunodeficiency virus. Analyses used Student’s t-test for continuous variables and Pearson’s chi-square test for categorical variables. Table 2. Differences in demographics and other diagnoses between intensive care unit patients with and without alcohol dependence Alcohol Dependence (n 1222) Demographics Age, median (interquartile range) Male gender, n (%) Nonwhite race, n (%) Emergency department admission, n (%) Other diagnoses, n (%) Liver disease HIV Trauma or burn Diabetes Infection or inflammation due to an in-dwelling device Nonseptic, non-HIV infection Cancer Leukemia or lymphoma Solid tumor Red cell transfusion No Alcohol Dependence (n 8759)

p Value

46 (40–53) 1004 (82.2) 625 (51.2) 1037 (84.9) 328 (26.8) 16 (1.3) 373 (30.5) 74 (6.1) 9 (0.7) 372 (30.4) 19 (1.6) 4 (0.3) 15 (1.2) 305 (25.0)

46 (3–60) 5598 (63.9) 4797 (54.8) 6636 (75.8) 272 (3.1) 165 (1.9) 3989 (45.5) 1557 (17.8) 53 (0.6) 1439 (16.4) 585 (6.7) 84 (1.0) 507 (5.8) 1819 (20.8)

.66 .001 .017 .001 .001 .16 .001 .001 .56 .001 .001 .022 .001 .001

HIV, human immunodeficiency virus. Analyses used Student’s t-test or Kruskal-Wallis test for continuous variables and Pearson’s chi-square test or Fisher’s exact test for categorical variables.

had coincident organ dysfunction and 227 (27.6%) had septic shock. Table 1 shows the unadjusted differences between patients with and without sepsis. Alcohol dependence was present in 1,222 (12.2%) of first-admissions. The majority had diagnoses of both alcoholism not in remission and alcohol withdrawal (n 620, 50.7%). The remainder had only alcoholism (n 439, 35.9%) or alcohol withdrawal (n 163, 13.3%). Table 2 illustrates unadjusted differences in demographics and comorbidities between
Crit Care Med 2007 Vol. 35, No. 2

those with and without alcohol dependence. Figure 2 shows the unadjusted associations between alcohol dependence and selected outcomes. Additionally, alcohol dependence was associated with more frequent organ failure (67.3% vs. 45.8%, p .001) and fewer hospital-free days (17.1 vs. 18.3, p .001). In a simple logistic regression, alcohol dependence was associated with sepsis (odds ratio [OR], 1.81; 95% confidence interval [CI], 1.50 –2.18; p .001). After we adjusted for factors previously shown

to be associated with sepsis, alcohol dependence remained significantly associated with sepsis (adjusted OR, 1.54; 95% CI, 1.25–1.91; p .001). For the final risk-adjusting model, a trauma or burn diagnosis was an additional confounder and red cell transfusion had a significant interaction with alcohol dependence (Table 3, p .004). To adjust for a general increased predilection to infections, we adjusted for the presence of infections other than sepsis and HIV. The association between alcohol dependence and sepsis remained statistically significant (Table 3). As with sepsis, alcohol dependence was significantly associated with septic shock in unadjusted analyses (OR, 1.75; 95% CI, 1.25–2.45; p .001). In the final risk-adjusting equation for septic shock, there was a statistically significant interaction between alcohol dependence and transfusion of red cells (p .001) that mirrored the association with sepsis (data not shown). Analysis using repeated measures for sepsis and septic shock produced similar results (data not shown). In unadjusted analyses, alcohol dependence was associated with higher odds of hospital mortality (OR, 1.28; 95% CI 1.04 – 1.57; p .022). Because we were interested in the association between alcohol dependence and mortality in those with sepsis, we examined the interaction between these diagnoses. Alcohol dependence was associated with a decreased hospital mortality among those without sepsis (OR, 0.94; 95% CI, 0.72–1.24) and with an increased mortality among septic patients (OR, 1.46; 95% CI, 1.01–2.11). After risk adjustment, the interaction term between alcohol dependence and sepsis was statistically significant (p .001). As shown in Table 4, there was also a significant interaction between alcohol dependence and liver disease (p .001). There was considerable variability in the association between alcohol dependence and hospital mortality, dependent on the presence of sepsis and/or liver disease. Those with all three diagnoses had almost six-fold higher adjusted odds of hospital mortality, relative to patients with none of these conditions.

DISCUSSION
In this retrospective cohort study, we found that one of every eight patients spending 1 day in the medical or surgical ICU at DHMC had alcohol dependence (alcoholism not in remission and/or alcohol withdrawal syndrome). Patients with alcohol dependence had higher odds of sepsis
347

Figure 2. Unadjusted differences in selected outcomes for intensive care unit patients with and without alcohol dependence. Alcohol dependence, sepsis, and septic shock were defined by codes from the International Classification of Diseases, 9th Revision, Clinical Modifications (see Material and Methods). p values from Pearson’s chi-square test. Table 3. Risk-adjusted association between alcohol dependence and sepsis Before adjustment for nonseptic, non-HIV infectionsa Alcohol dependence Patients not receiving red cell transfusion Patients receiving red cell transfusion Red cell transfusion Patients without alcohol dependence Patients with alcohol dependence After adjustment for nonseptic, non-HIV infectionsb Alcohol dependence Patients not receiving red cell transfusion Patients receiving red cell transfusion Red cell transfusion Patients without alcohol dependence Patients with alcohol dependence

1.91 (1.49–2.44) 0.92 (0.65–1.31) 3.23 (2.73–3.83) 1.57 (1.08–2.27) 1.74 (1.35–2.23) 0.86 (0.60–1.22) 2.86 (2.40–3.40) 1.41 (0.96–2.07)

HIV, human immunodeficiency virus. a Adjusted for age, race, gender, liver disease, HIV, leukemia or lymphoma, solid organ cancer, infection due to an indwelling device, red blood cell transfusion, and trauma or burns. Interaction between alcohol dependence and red cell transfusion, p .001. c statistic 0.71. bAdjusted for age, race, gender, liver disease, HIV, leukemia or lymphoma, solid organ cancer, infection due to an indwelling device, red blood cell transfusion, trauma or burns, and nonseptic, non-HIV infections. Interaction between alcohol dependence and red cell transfusion, p .001. c statistic 0.76. Table 4. Risk-adjusted association between alcohol dependence and hospital mortality Alcohol dependence Patients without sepsis Patients without liver disease Patients with liver disease Patients with sepsis Patients without liver disease Patients with liver disease Sepsis Patients without alcohol dependence Patients with alcohol dependence Liver disease Patients without alcohol dependence Patients with alcohol dependence

0.30 (0.20–0.46) 0.86 (0.52–1.42) 0.81 (0.48–1.37) 2.31 (1.26–4.24) 1.72 (1.35–2.20) 4.66 (2.81–7.72) 1.71 (1.18–2.49) 4.87 (2.99–7.93)

Odds ratios for hospital mortality (95% confidence interval) after adjustment for number of organ failures, age, human immunodeficiency virus (HIV), red blood cell transfusion, leukemia or lymphoma, solid organ cancer, trauma or burn diagnoses, nonseptic, non-HIV infections, emergency department admission, and diabetes mellitus. Interaction between alcohol dependence and sepsis, p .001. Interaction between alcohol dependence and liver disease, p .001. c statistic 0.88.

and septic shock. They also had higher hospital mortality and fewer hospital-free days. After risk adjustment, alcohol dependence was independently associated
348

with odds of sepsis, septic shock, and hospital mortality. Alcohol Use Disorders and Sepsis. Several studies have examined the associa-

tion between alcohol use disorders and the incidence of severe infections that could reasonably be considered sepsis. This includes reports of increased risk for severe bloodstream infections (23) and higher severity of illness, morbidity, and mortality among patients with community-acquired pneumonia (17, 18). In patients with at-risk diagnoses, alcohol use disorders were independently associated with the development of the acute respiratory distress syndrome and increased hospital mortality for those who developed acute respiratory distress syndrome (19 –21). We found that alcohol dependence was independently associated with higher odds of sepsis. A general predilection to infection mediated some of this association, but alcohol dependence remained significantly associated with higher odds of sepsis. This effect was modified by red blood cell transfusions. Alcohol dependence increased the odds of sepsis by 74% among those not receiving a transfusion and was associated with a trend toward decreased odds of sepsis in transfused patients. We caution that this interaction does not indicate that alcohol dependence is protective in the context of red blood cell transfusions. Since the confidence interval for this stratum includes the null, concluding a benefit of alcohol dependence in those receiving red cell transfusions would be subject to a type 1 error. The mechanism of the modification of the association between alcohol dependence, red cell transfusions, and sepsis remains to be described. In analyses for hospital mortality, liver disease and sepsis modified the association between alcohol dependence and death. The greatest risk-adjusted odds of death were observed in patients having all three diagnoses. Interestingly, among nonseptic patients without liver disease, those with alcohol dependence had 71% lower odds of death than those without alcohol dependence. Alcohol-dependent patients without liver disease and sepsis appear to be a relatively low-risk population of ICU patients, consistent with prior studies indicating that alcohol use disorders may trigger ICU admission for those with an otherwise low severity of illness (24). Possible Mechanisms of the Association Between Alcohol Dependence and Sepsis. Multiple animal and human studies have demonstrated abnormal immunity as a result of alcohol exposure (14, 25–28). This includes abnormalities in innate and adaptive immunity; cellular
Crit Care Med 2007 Vol. 35, No. 2

and humoral responses; the functioning of neutrophils, monocytes, macrophages and lymphocytes; and cytokine and chemokine profiles. Such differences could lead to an increased predilection to infection and, once established, an increased risk of systemic complications. Alteration of glutathione homeostasis may be an additional mechanism by which alcohol abuse predisposes septic patients to organ failure (29, 30). Limitations. As this was a retrospective cohort study, we must rely on the validity, accuracy, and comprehensiveness of the collected data. UHC is a data set that relies on information from billing and administrative sources. Our ability to explore the association between alcohol dependence and the development of severe sepsis was limited because alcohol, by itself, can lead to organ failure. This prevented us from confirming the association between alcohol use disorders and acute respiratory distress syndrome. Instead of prospectively identifying diagnoses of interest, we relied on validated diagnostic codes for the designation of sepsis and various comorbidities. There are published combinations of ICD-9 codes for alcohol-related diagnoses that include diagnoses associated with alcohol intoxication, abuse, and dependence (18, 31). However, sensitivity is moderate (77%) when validated against medical records, a technique shown by others to underestimate the prevalence of alcohol abuse (32). Such an approach would also include subjects with acute alcohol use disorders rather than only those with chronic dependence. There remains the possibility of bias due to misclassification or detection, but we doubt a systematic variation in the findings of an alcohol dependence diagnosis based on the coincident diagnosis of sepsis. A prospective study validating ICD-9 codes against an accurate measure of alcohol dependence would be valuable for future examination of the association between alcohol use disorders and critical illness. These data represent admissions to two ICUs at a single “safety net” hospital. The ability to generalize our findings to hospitals with other case mixes is unknown. We are uncertain if our findings are relevant to short-stay ICU patients and non-ICU patients. Although we adjusted for factors previously shown to be associated with increased risk of sepsis, the data set limited our ability to assess other possible covariates. We included alcohol-associated conditions, such as cirrhosis and cancer, in our risk adjustment,
Crit Care Med 2007 Vol. 35, No. 2

but other factors found at higher rates among alcohol abusers, such as smoking and malnutrition, could not be included. We also cannot comment on the association between alcohol use disorders other than alcohol dependence and the outcomes under study. Due to limitations in the data, we were unable to use a physiologically based severity of illness score. Instead, we included known factors associated with sepsis, explored additional available factors, and examined the effect of repeated admissions on our findings. Our analyses also showed that consideration of interactions between covariates is important to provide a complete picture of the association between alcohol dependence and sepsis.

9.

10.

11.

12.

13.

CONCLUSIONS
In this retrospective cohort study of adult ICU patients, alcohol dependence (alcoholism not in remission and/or alcohol withdrawal syndrome) was independently associated with sepsis, septic shock, and hospital mortality. An increased predisposition to infection mediated some of this association. Elucidation of the mechanism of increased risk of sepsis for patients with alcohol dependence may help to define pathologic mechanisms of sepsis and sepsis-induced organ dysfunction and death.
14.

15.

16. 17.

18.

REFERENCES
1. Medical consequences of alcohol abuse. Alcohol Res Health 2000; 24:27–31 2. Mokdad AH, Marks JS, Stroup DF, et al: Actual causes of death in the United States, 2000. JAMA 2004; 291:1238 –1245 3. Harwood H: Updating Estimates of the Economic Costs of Alcohol Abuse in the United States: Estimates, Update Methods and Data. Bethesda, MD, The Lewin Group for the National Institute on Alcohol Abuse and Alcoholism, 2000 4. Marik P, Mohedin B: Alcohol-related admissions to an inner city hospital intensive care unit. Alcohol Alcohol 1996; 31:393–396 5. Mostafa SM, Murthy BV: Alcohol-associated admissions to an adult intensive care unit: An audit. Eur J Anaesthesiol 2002; 19: 193–196 6. Hoyert DL, Arias E, Smith BL, et al: Deaths: Final data for 1999. Natl Vital Stat Rep 2001; 49:1–113 7. Angus DC, Linde-Zwirble WT, Lidicker J, et al: Epidemiology of severe sepsis in the United States: Analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001; 29:1303–1310 8. Martin GS, Mannino DM, Eaton S, et al: The epidemiology of sepsis in the United States 19.

20.

21.

22.

23.

24.

25.

from 1979 through 2000. N Engl J Med 2003; 348:1546 –1554 Foreman MG, Mannino DM, Moss M: Cirrhosis as a risk factor for sepsis and death: Analysis of the National Hospital Discharge Survey. Chest 2003; 124:1016 –1020 Witt DJ, Craven DE, McCabe WR: Bacterial infections in adult patients with the acquired immune deficiency syndrome (AIDS) and AIDS-related complex. Am J Med 1987; 82: 900 –906 Williams MD, Braun LA, Cooper LM, et al: Hospitalized cancer patients with severe sepsis: Analysis of incidence, mortality, and associated costs of care. Crit Care 2004; 8:R291–R298 Groeger JS, Lucas AB, Thaler HT, et al: Infectious morbidity associated with long-term use of venous access devices in patients with cancer. Ann Intern Med 1993; 119:1168 –1174 Raghavan M, Marik PE: Anemia, allogenic blood transfusion, and immunomodulation in the critically ill. Chest 2005; 127:295–307 Diaz LE, Montero A, Gonzalez-Gross M, et al: Influence of alcohol consumption on immunologic status: A review. Eur J Clin Nutr 2002; 56:S50 –S53 Fernandez-Sola J, Junque A, Estruch R, et al: High alcohol intake as a risk and prognostic factor for community-acquired pneumonia. Arch Intern Med 1995; 155:1649 –1654 Hudolin V: Tuberculosis and alcoholism. Ann N Y Acad Sci 1975; 252:353–364 Ruiz M, Ewig S, Torres A, et al: Severe community-acquired pneumonia. Risk factors and follow-up epidemiology. Am J Respir Crit Care Med 1999; 160:923–929 Saitz R, Ghali WA, Moskowitz MA: The impact of alcohol-related diagnoses on pneumonia outcomes. Arch Intern Med 1997; 157: 1446 –1452 Moss M, Bucher B, Moore FA, et al: The role of chronic alcohol abuse in the development of acute respiratory distress syndrome in adults. JAMA 1996; 275:50 –54 Moss M, Parsons PE, Steinberg KP, et al: Chronic alcohol abuse is associated with an increased incidence of acute respiratory distress syndrome and severity of multiple organ dysfunction in patients with septic shock. Crit Care Med 2003; 31:869 – 877 Hudson LD, Milberg JA, Anardi D, et al: Clinical risks for development of the acute respiratory distress syndrome. Am J Respir Crit Care Med 1995; 151:293–301 Deyo RA, Cherkin DC, Ciol MA: Adapting a clinical comorbidity index for use with ICD9-CM administrative databases. J Clin Epidemiol 1992; 45:613– 619 Laupland KB, Gregson DB, Zygun DA, et al: Severe bloodstream infections: A populationbased assessment. Crit Care Med 2004; 32: 992–997 Rosenthal GE, Sirio CA, Shepardson LB, et al: Use of intensive care units for patients with low severity of illness. Arch Intern Med 1998; 158:1144 –1151 Gamble L, Mason CM, Nelson S: The effects

349

26.

27.

28.

29.

30.

31.

32.

of alcohol on immunity and bacterial infection in the lung. Med Mal Infect 2006; 36: 72–77 Happel KI, Nelson S: Alcohol, immunosuppression, and the lung. Proc Am Thorac Soc 2005; 2:428 – 432 Pavia CS, La Mothe M, Kavanagh M: Influence of alcohol on antimicrobial immunity. Biomed Pharmacother 2004; 58:84 – 89 Szabo G: Consequences of alcohol consumption on host defence. Alcohol Alcohol 1999; 34:830 – 841 Moss M, Guidot DM, Wong-Lambertina M, et al: The effects of chronic alcohol abuse on pulmonary glutathione homeostasis. Am J Respir Crit Care Med 2000; 161:414 – 419 Guidot DM, Modelska K, Lois M, et al: Ethanol ingestion via glutathione depletion impairs alveolar epithelial barrier function in rats. Am J Physiol Lung Cell Mol Physiol 2000; 279:L127–L135 Dufour MC, Caces MF: Epidemiology of the medical consequences of alcohol. Alcohol Health Res World 1993; 17:265–271 Umbricht-Schneiter A, Santora P, Moore RD: Alcohol abuse: Comparison of two methods for assessing its prevalence and associated morbidity in hospitalized patients. Am J Med 1991; 91:110 –118

APPENDIX ICD-9 and CPT [Current Procedural Terminology] Codes
Alcohol dependence Alcohol withdrawal syndrome 291.0 —Alcohol withdrawal delirium 291.3—Alcohol-induced psychotic disorder with hallucinations 291.5—Alcohol-induced psychotic disorder with delusions 291.81—Alcohol withdrawal psychosis 291.89 —Other specified alcoholinduced mental disorders 291.9 —Unspecified alcohol-induced mental disorders Alcoholism, not in remission 303.90 —Other and unspecified alcohol dependence, unspecified drinking behavior 303.91—Other and unspecified alcohol dependence, continuous drinking behavior 303.92—Other and unspecified alcohol dependence, episodic drinking behavior

Sepsis 038 —septicemia 112.5— disseminated Candidiasis 112.81—-candidal endocarditis 117.9 — disseminated fungal infection 790.7— bacteremia Shock 458.0 — orthostatic hypotension 458.8 — other specified hypotension 458.9 — unspecified hypotension 785.5—shock without mention of trauma 796.3—nonspecific low blood pressure reading Solid tumor 140 –149 —malignant neoplasm of the lip, oral cavity, and pharynx 150 –159 —malignant neoplasm of the digestive organs and peritoneum 160 —malignant neoplasm of nasal cavities, middle ear, and accessory sinuses 161–165—malignant neoplasm of the larynx, trachea, bronchus, lung, pleura, and intrathoracic organs 170 —malignant neoplasm of bone and articular cartilage 171—malignant neoplasm of connective and other soft tissue 172—malignant melanoma 174 –175—malignant neoplasm of the breast 176 —Kaposi sarcoma 179 –184 —malignant neoplasm of the uterus, cervix, placenta, ovary, and female genital organs 185–187—malignant neoplasm of prostate, testis, penis, and male genital organs 188 –189 —malignant neoplasm of bladder, kidney, and urinary tract 190 —malignant neoplasm of the eye 191–192—malignant neoplasm of brain and nervous system 193–194 —malignant neoplasm of the thyroid and other endocrine organs 195—malignant neoplasm of illdefined sites 196 —metastatic tumors of lymph nodes 197—metastatic cancer of respiratory and gastrointestinal system 198 —metastatic cancer of other specified sites

199 —malignant neoplasm without specification of site Leukemia or lymphoma 200 —lymphosarcoma and reticulosarcoma 201—Hodgkin disease 202— other malignant neoplasms of lymphoid and histiocytic tissue 203—multiple myeloma and immunoproliferative neoplasms 204 —lymphoid leukemia 205—myeloid leukemia 206 —monocytic leukemia 207— other specified leukemia 208 —leukemia of unspecified cell type Liver disease 456 —varicose veins of other sites 571.2—alcoholic cirrhosis of the liver 571.4 — chronic hepatitis 571.5— cirrhosis of the liver without mention of alcohol 572.2— hepatic coma 572.3—portal hypertension 572.4 — hepatorenal syndrome 572.8 — other sequelae of chronic liver disease HIV 042— human immunodeficiency virus (HIV) disease 043—HTLV-III [human T-cell lymphotrophic virus type III] conditions 044 — other HTLV-III infections Red cell transfusion 99.00 —perioperative autologous transfusion of whole blood or blood components 99.01— exchange transfusion 99.02—transfusion of previously collected autologous blood 99.03— other transfusion of whole blood 99.04 —transfusion of packed cells Infection/inflammation due to a device 996.6 —infection and inflammatory reaction due to internal prosthetic device, implant, or graft Descriptions of codes are from the UHC CDP Online Database (www.uhc. edu). Codes are ICD-9 codes, except for red cell transfusion (CPT codes).

350

Crit Care Med 2007 Vol. 35, No. 2


				
DOCUMENT INFO
Categories:
Stats:
views:498
posted:8/1/2009
language:English
pages:6
Description: Sepsis mortality.