An Evaluation of Treatment Patterns and Associated Outcomes Among Adult Hospitalized Patients With Lower-Risk Community-Acquired Complicated Intra-abdominal Infections: How Often Are Expert Guidelines Followed?

BACKGROUND: Expert guidelines discourage use of antipseudomonal β-lactams and fluoroquinolones in lower-risk patients with community-acquired complicated intra-abdominal infection (CA cIAI). Compliance with these recommendations across US hospitals is unclear. This study sought to determine treatment patterns and associated outcomes among adult hospitalized lower-risk patients with CA cIAI.
METHODS: A study using data from the Premier Healthcare Database (10/2015-12/2017) was performed. Inclusion criteria: age ≥18 years; hospitalized; had a cIAI at admission; and received antibiotics within the first 4 hospital days. Patients were excluded if they were high risk, were transferred from another health care facility, had a recent hospital admission, or received dialysis within 30 days of admission. Empiric antibiotic treatment patterns and associated outcomes were quantified.
RESULTS: Overall, 46 722 (66%) patients with cIAIs met the lower-risk CA IAI study criteria. Among lower-risk CA IAI patients, the mean (SD) age was 53.4 (18.2) years, and 71% had a Charlson Comorbidity Index score of 0. The most common diagnosis was acute appendicitis with peritonitis (59.7%). Among lower-risk CA IAI patients, 54% received piperacillin/tazobactam, 20% received a fluoroquinolone (FQ), 11% received ceftriaxone, and 7% received ampicillin/sulbactam. Overall, the median hospital length of stay was 4 days and median costs were $12 345 USD. Nearly 90% of patients were discharged home, and <1% died. Outcomes were similar across all empiric treatments received.
CONCLUSIONS: Overuse of antipseudomonal β-lactams and fluoroquinolones was commonplace among lower-risk CA IAI patients. These findings can serve as the basis for an antimicrobial stewardship initiative in hospitals aspiring to reduce the use of broad-spectrum antibiotics.

Antimicrobial therapy plays an integral role in the management of patients with complicated intra-abdominal infections (cIAIs). Empiric treatment selection is based on patients’ location before cIAI and background medical conditions, severity of infection, anatomical site of infection, and antibiotic resistance rates at the local health care institution [1, 2]. Broadly, treatment selection is based on whether the cIAI is community-acquired (CA) or health care–associated (HA) and if the patient is characterized as having a lower or higher risk for treatment failure or death. Most patients with cIAI meet the CA and lower-risk classifications [2]. Among patients with lower-risk CA cIAI, expert guidelines recommend narrower-spectrum antimicrobial agents with activity against the “common gram-negative Enterobacteriaceae, aerobic streptococci, and obligate anaerobic microorganisms.” Per the Surgical Infection Society Revised Guidelines on the Management of IAI [2], recommended empiric antimicrobial regimens for patients with lower-risk CA cIAI include ertapenem, moxifloxacin, cefotaxime, or ceftriaxone plus metronidazole, and in penicillin-allergic patients, ciprofloxacin (levofloxacin in formulary fluoroquinolone) plus metronidazole. Carbapenems (except etrapenem) and piperacillin/tazobactam (TZP) are discouraged in adult lower-risk patients with CA cIAI to avoid excessive use and potential promotion of resistance [1, 2]. Fluoroquinolones are also not recommended for use in institutions with high rates of fluoroquinolone resistance among common gram-negative Enterobacteriaceae. There are also growing safety concerns with use of fluroquinolones. The US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) have updated the labeling of all fluoroquinolones, advising of the serious risk of multiple disabling and potentially irreversible adverse reactions associated with their use [3, 4]. Most recently, the FDA and EMA updated the labeling of all fluoroquinolones to include the increased risk of aortic aneurysm associated with their use and recommending prescription of fluoroquinolones to patients only when no other treatment options are available [3, 4].

While treatment recommendations are well delineated in the guidelines, there are scant data on how well US hospitals have adopted these recommendations [5]. More importantly, it is not clear how often carbapenems, TZP, cefepime, and fluoroquinolones are used in clinical practice among lower-risk patients with CA cIAIs. There are also limited real-world data on the outcomes associated with the antibiotics used in the treatment of adult hospitalized lower-risk patients with CA cIAI. Given these gaps in the literature, the intent of this descriptive study was to examine antibiotic treatment patterns and associated outcomes among adult hospitalized lower-risk patients with CA cIAI across US hospitals. Emphasis was placed on quantifying use of antipseudomonal carbapenems, TZP, cefepime, and fluoroquinolones among adult hospitalized lower-risk patients with CA cIAI.

METHODS

Study Design

A retrospective observational study was conducted to examine treatment patterns and associated outcomes of lower-risk adult patients with CA cIAI across US hospitals. Data for the study are from the Premier Healthcare Database, which currently contains data from >730 million patient encounters [6]. The Premier Healthcare Database contains data from standard hospital discharge files, including a patient’s demographic and disease states. In addition to the data elements available in most of the standard hospital discharge files, the Premier Healthcare Database also contains a date-stamped log of billed items, including procedures, medications, and laboratory, diagnostic, and therapeutic services at the individual patient level. Information on hospital characteristics, including geographic location, bed size, and teaching status, is also available. However, no clinical laboratory values are available in the database. Preliminary comparisons between patient and hospital characteristics for the hospitals that submit data to Premier Healthcare Database and those of the probability sample of hospitals and patients selected for the National Hospital Discharge Survey (NHDS) suggest that the patient populations are similar with regard to patient age, gender, length of stay, mortality, primary discharge diagnosis, and primary procedure groups. The database was fully de-identified and compliant with the Health Insurance Portability and Accountability Act of 1996 (HIPAA); as such, no special permission was required to review patient records and extract the data. Given the de-identified and retrospective nature of the data, as well as the observational study design, written patient consent was neither required nor sought.

Study Population

This study included all inpatient discharges from patients aged 18 years or older with evidence of a cIAI between 10/2015 and 12/2017 [7, 8]. The first qualifying cIAI discharge during the study period was defined as the index cIAI admission.

An inpatient hospitalization for patients aged 18 years or older, discharged between 10/2015 and 12/2017.

The first qualifying hospitalization with evidence of a cIAI was flagged as the index hospitalization.

For patients with multiple cIAI admissions, only the first cIAI was considered.

Evidence of cIAI defined by algorithms based on diagnosis or procedure codes (Appendix A).

Primary cIAI diagnosis and a cIAI surgical procedure or a secondary cIAI diagnosis and cIAI surgical procedure within 5 days of admission.

Received an antibiotic within the first 4 hospital days.

Met criteria for high-risk patient (sepsis, severe sepsis, septic shock; ≥3 components of sepsis; or ≥2 physiologic risk factors) at admission (Appendix B) [2].

Had a hospital admission in past 30 days.

Health care facility point of origin (transferred from another hospital or health care facility).

Died or was discharged alive on the index date.

Received any dialysis within 30 days before index admission.

Treatment Classification

Patients were assigned into non–mutually exclusive cohorts based on antibiotics received during the first 4 days of hospitalization. Empiric treatment regimens received during the first 4 days of hospitalization were determined. We also calculated the number of patients who received 1 of the following antibiotics during the first 4 days of hospitalization: TZP, meropenem, cefepime, and fluoroquinolones.

Baseline Covariates

Patients’ demographic and clinical characteristics were based on available information during the qualifying admission period. Patient-level covariates included in the analysis were demographics, comorbidities, Charlson Comorbidity Index (CCI) scores [9], cIAI infection designation, and antibiotics received during qualifying admission. Hospital-level covariates included region, population served, teaching status, and hospital size. Outcome measures included duration of antibiotic therapy, hospital length of stay (LOS), hospital costs, in-hospital mortality, discharge destination (eg, home, long-term care facility, skilled nursing facility, hospice), and 30-day readmissions postdischarge.

Statistical Methods

Unadjusted descriptive statistics were used to characterize the patient population. Patient demographics, clinical conditions, hospital characteristics, and outcomes were examined, and summary statistics were reported. Data measured on a continuous scale were expressed as median and interquartile range (IQR). Dichotomous and categorical data were expressed as counts and percentages of patients in the categories. Outcomes were stratified by empiric treatment received, but no inferential statistics were performed as this was a descriptive study. Analyses were conducted using SAS software, version 9.3 (SAS Institute, Cary, NC, USA).

RESULTS

During study period, 77 663 patients had a qualifying hospitalization with evidence of a cIAI, representing ~2% of all admissions during study time frame. Among the 77 663 patients, 30 941 did not meet the study criteria, resulting in a final study population of 46 722 lower-risk patients with CA cIAIs. Baseline hospital- and patient-level characteristics of the lower-risk cIAI study population are shown in Table 1. Most lower-risk CA cIAI patients received care in hospitals classified as urban (88.4%) and nonteaching (62.7%). The largest category of hospital size was >500 beds (29.2%), and the South was the region with the most patients (41.7%). The mean (SD) age was 54 (18) years, and 52% were male. Most patients were white (76.3%), the median (IQR) CCI score was 0 (0–1), and 71% of patients had a CCI score of 0. Chronic lung diseases (12.9%) and diabetes without complications (13.3%) were the only comorbidities present in >10% of the study population. Acute appendicitis with peritonitis was the most common diagnosis (59.7%), followed by peritoneal abscess (9.9%), non-trauma-related perforation of intestine (8.9%), and fistula of intestine (5.1%).

Table 1.

Baseline Characteristics of Included Patients (n = 46 722)

Age, mean (SD), y	54 (18)
Sex, male	24 657 (52)
Race
White	35 646 (74.3)
Black or African American	5746 (12.3)
Other	5330 (11.4)
Charlson Comorbidity Index, median (IQR)	0 (0–1)
Myocardial infarction	1510 (3.2)
Congestive heart failure	2384 (5.1)
Peripheral vascular disease	618 (1.3)
Dementia	874 (1.9)
Cerebrovascular disease	489 (1.0)
Chronic lung disease	6042 (12.9)
Connective tissue disease	827 (1.8)
Ulcer	3685 (7.9)
Chronic liver disease	2353 (5.0)
Hemiplegia	205 (0.4)
Moderate or severe kidney disease	2838 (6.1)
Diabetes without complications	6232 (13.2)
Diabetes with complications	1433 (3.1)
Tumor	2849 6.1)
Malignant tumor, metastasis	1287 (2.8)
AIDS	97 (0.2)
Complicated intra-abdominal infection diagnosis
Acute appendicitis with peritonitis	28 004 (59.7)
Peritoneal abscess	4624 (9.9)
Perforated intestine	4142 (8.9)
Fistula of intestine	2406 (5.1)
Peritonitis (unspecified)	1901 (5.1)
Hospital-level covariates
Hospital region
South	19 481 (41.7)
West	10 670 (22.8)
Midwest	9072 (19.4)
Northeast	7499 (16.1)
Population served
Urban	41 304 (88.4)
Rural	5418 (11.6)
Teaching hospital	29 315 (62.7)
Nonteaching hospital	17 407 (37.3)
No. of hospital beds
000–099	3092 (6.6)
100–199	7428 (15.9)
200–299	9228 (19.8)
300–399	8060 (17.3)
400–499	5263 (11.3)
500+	13 651 (29.2)

Data are presented as No. (%) unless otherwise indicated.

Abbreviation: IQR, interquartile range.

The most frequently used antibiotics during the first 4 days of hospitalization are shown in Figure 1. Piperacillin/tazobactam (54%) was the most commonly used antibiotic. Fluroquinolone-containing regimens (ciprofloxacin and levofloxacin) were used in 20% of the study population. Of the 9339 patients who received a fluoroquinolone, 8066 (86.3%) also received metronidazole. Approximately 11% of the study population received ceftriaxone, and 7.5% received ampicillin/sulbactam. Of the 5006 who received ceftriaxone, 3959 (79.1%) also received metronidazole. Cefepime and meropenem were prescribed less frequently (each prescribed in 3% of patients). In total, 8% received ≥2 of the following agents during the first 4 days of hospitalization: TZP, meropenem, cefepime, and fluoroquinolone.

Figure 1.

Most common empiric antibiotics received among patients with lower-risk complicated intra-abdominal infections.

Overall, the median (IQR) LOS was 4 (2–8) days, and the median (IQR) total hospital cost was $12 345 ($8447–$20 399) US dollars. Less than 1% of patients died during their hospitalization, and most were discharged home (89.0%) or transferred to another hospital (9.5%). Approximately 10% of patients had a hospital re-admission within 30 days of discharge. Outcomes of patients who received a TZP-, meropenem-, cefepime-, or fluroquinolone-containing regimen are shown in Table 2. Although this study was not designed to compare outcomes across regimens received and no inferential statistics were performed, hospital LOS, total costs, discharge destination, and 30-day readmission rates were largely similar across regimens, with the exception of those who received meropenem or cefepime. Closer examination of these patients indicated that they were typically older, had more comorbid conditions, and had higher CCI scores (data not shown). Outcomes were also alike among lower-risk CA cIAI patients who received 0, 1, or ≥2 of the following antibiotics: TZP, meropenem, cefepime, and fluoroquinolone(s) (Table 3).

Table 2.

Outcomes by Empiric Therapy Received

	Overall	TZP	FQ	CR0	SAM	MEM	FEP
No. of patients (%)	46 722 (100)	25 070 (54.7)	9339a (20.0)	5006b (10.7)	3492 (7.4)	1308 (2.8)	1331 (2.8)
LOT, median (IQR), d	5 (3–8)	4 (3–6)	3 (1–5)	2 (1–5)	3 (1–4)	5 (2–8)	3 (1–5)
LOS, median (IQR), d	4 (2–8)	5 (3–8)	5 (3–8)	5 (3–9)	4 (2–7)	7 (4–12)	7(4–12)
Hospital costs, median (IQR), $	12 345 (8447–20 399)	12 377 (8591–19 748)	12 823 (8829–20 836)	13 995 (9336–22 470)	11 723 (8409–18 509)	18 256 (11 680–30 724)	18 388 (11 047–31 101)
30-d hospital re-admission, %	9.9	9.7	9.4	9.9	8.9	12.8	12.2
Hospital discharge location, %
Home	88.9	89.5	87.1	85.0	91.7	78.0	76.5
Hospital	9.5	8.9	11.3	12.9	6.9	18.5	19.6
Death	0.9	0.9	0.9	1.1	0.8	2.1	2.4
Hospice	0.6	0.6	0.7	1.0	0.5	1.38	1.4

Abbreviations: CR0, ceftriaxone; FEP, cefepime; FQ, fluoroquinolones; IQR, interquartile range; LOS, hospital length of stay; LOT, length of therapy; MEM, meropenem; SAM, ampillin/sulbactam; TZP, piperacillin/tazobactam.

a8066 also received metronidazole.

b3959 also received metronidazole.

Table 3.

Outcomes by Receipt of 1, 2, and ≥2 of the Following Antibiotics: Piperacillin/Tazobactam, Meropenem, Cefepime, Ciprofloxacin, and Levofloxacin

	None	1	≥2
No. of patients	7028	23 827	15 867
LOS, median (IQR), d	4 (2–8)	4 (2–7)	5 (3–9)
Hospital costs, median (IQR), $	11 440 (7723–21 159)	11 781 (8211–19 079)	13 664 (9295–22 052)
30-d hospital re-admission, %	10.9	9.8	9.7
Hospital discharge location, %
Home	89.4	90.1	86.8
Hospital	8.9	8.5	11.3
Death	1.0	0.8	1.1
Hospice	0.7	0.5	0.8

Abbreviations: IQR, interquartile range; LOS, length of stay.

DISCUSSION

Overall, the findings from this study suggest that use of non-guideline-concordant therapies for patients with lower-risk CA cIAIs is commonplace. Empiric therapy with TZP or fluoroquinolones occurred in >75% of patients meeting the criteria for lower-risk CA cIAI employed in this study. The rationale for discouraging the empiric use of these agents in low-risk CA cIAI relates to their spectrum of activity [1, 2]. The concerns with fluoroquinolones stem from a lack of antibacterial activity against common gram-negative pathogens associated with cIAI. Rates of fluoroquinolone resistance among Escherichia coli, the most common cIAI pathogen, now exceed 25% in most regions in the United States, limiting their utility as an effective first-line empiric option [10]. In contrast to fluoroquinolones, TZP is an overly broad-spectrum antipseudomonal β-lactam therapy for lower-risk CA cIAI patients. One of the primary reasons to use TZP is for patients with suspected or documented P. aeruginosa infections. Among lower-risk patients with CA cIAI, P. aeruginosa represented <5% of culture-positive cIAI patients in a recently published multicenter observational study [11]. It is important to note that cultures are not often collected in patients with lower-risk CA cIAIs, and the point estimate of 5% is likely conservative. Furthermore, the need to even treat P. aeruginosa in cIAI patients without high acuity of illness is questionable, as studies have shown successful outcomes in patients who did not receive an antipseudomonal agent [1, 2, 12–14].

Despite being overly broad, the extensive use of TZP in lower-risk CA cIAI patients suggests that its use is not considered problematic by many clinicians. While certain antibiotics like carbapenems, third- and fourth-generation cephalosporins, and fluoroquinolones are generally considered the agents that pose the greatest risks, use of all broad-spectrum antibiotics, including TZP, increases the risk of CDI and antibiotic resistance [15, 16]. In a recent case–control study of surgical trauma patients, CDI cases were more likely to have received TZP relative to CDI-negative controls (odds ratio [OR], 2.4; 95% CI, 1.3–4.5) [17]. Similarly, in a study involving 64 US academic medical centers, use of β-lactam/β-lactamase inhibitor combinations (of which 80% was due to TZP use) was associated with an increased odds of acquiring an HA-CDI (OR, 1.49; 95% CI, 1.36–1.64) [18]. Interestingly, the odds of acquiring an HA-CDI were found to be relatively comparable among patients who received third- or fourth-generation cephalosporins (OR, 1.75; 95% CI, 1.62–1.89), carbapenems (OR, 1.60; 95% CI, 1.44–1.79), and β-lactam/β-lactamase inhibitor combinations (OR, 1.49; 95% CI, 1.36–1.64). Clinical observational studies also suggest that prior exposure to TZP, albeit to a lesser extent than carbapenems, increases the risk of acquiring a carbapenem-resistant gram-negative infection [19-21]. In support of these clinical observations, a recent study of ICU patients indicated that TZP contributes to microbiota disruption and leads to increased carbapenem-resistant P. aeruginosa acquisition [22]. There are also data that correlate TZP use with the increased rates of TZP resistance in E. coli and Klebsiella sp. [23]. Additionally, Teshome and colleagues demonstrated that each additional day of treatment with TZP was associated with an 8% risk of new resistance development among patients with severe sepsis or septic shock [24].

Although this study was not designed to compare outcomes by treatment received and no inferential statistics were performed, another interesting observation from this study was that outcomes were largely similar across the varying empiric treatment regimens used in lower-risk cIAI patients. This observation needs to be interpreted cautiously, as no inferential statistics were performed, but it potentially highlights that there may be an opportunity to use more narrow-spectrum antibiotic therapies and limit the use of fluoroquinolones and antibiotics with antipseudomonal activity. Studies demonstrate that shifts in antibiotic usage patterns, regardless of patient population, are best accomplished through the development of interdisciplinary guidelines and staff education [25-28]. Site-of-care clinical pathways, especially those built into electronic medical record support systems, empower antibiotic stewardship programs to ensure that the appropriate therapies are empirically initiated and provide an avenue for them to affect therapy changes when performing perspective audit and feedback if guidelines are not followed upfront [25].

There are several things to consider when interpreting the findings of this study. This was a descriptive study, and its intent was simply to highlight the proportion of adult hospitalized lower-risk patients with cIAI who received a non-guideline-concordant therapy. This study was not designed to compare outcomes by treatment received, nor does it control for the multitude of factors that influence diverse outcomes like readmissions or LOS. To properly compare outcomes across treatment groups, additional analyses that account for baseline differences between groups would have been needed. Laboratory or other electronic medical record information was not available in the Premier Healthcare Database. We therefore had to use International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM), diagnosis codes to derive the cIAI cohort and the disease severity and comorbidity measures used in this study. It is quite possible that not all clinical conditions were coded and some higher risk patients and HA cIAIs were misclassified as lower risk and CA cIAIs, respectively. To minimize major “risk” classification errors (ie, classifying lower-risk patients as higher-risk), we developed comprehensive algorithms to fully capture patients with sepsis and physiologic risk factors. We also implemented several safeguards to ensure that HA cIAIs were not classified as CA. In addition to laboratory and electronic medical data, microbiologic data were not included in this study. It is quite possible that patients classified as lower risk CA cIAI received a board-spectrum antibiotic to treat a suspected or documented antibiotic-resistant pathogen like an extended spectrum β-lactamase (ESBL)–producing Enterobacteriaceae recovered on culture. Of note, the SIS Surgical Infection Society guidelines do not recommend routinely obtaining peritoneal fluid cultures in lower-risk patients with CA IAI for the purposes of guiding antimicrobial therapy, as studies suggest that cultures rarely, if ever, provide information useful to the clinician [2, 29]. We also did not have allergy information, and it is possible that some of the patients who received a fluoroquinolone were intolerant to β-lactams. Conservatively, even if the approach employed in this study misclassified 30%–50% of patients, the findings would still indicate that the majority of lower-risk patients with CA received non-guideline-concordant therapy. Of note, sepsis, severe sepsis, and septic shock are well-defined codes in the ICD-10-CM, and it unlikely that many patients with any of these conditions would have been missed as their codes are critical for reimbursement [30]. In support of this, the mortality rates and hospital LOS among lower-risk patients with CA cIAIs were very low, suggesting that the algorithm for classifying patients was fairly accurate. However, the algorithms used in this study were not designed or intended to be treatment prediction tools to define therapy in adult patients who presented to the hospital with a cIAI. In order for this to serve as a prediction tool, it will need to be prospectively validated across multiple institutions using data from the medical record.

In conclusion, overuse of non-guideline-concordant broad-spectrum antipseudomonal antibiotics, namely TPZ and fluoroquinolones, was found to be commonplace among lower-risk patients with CA cIAIs. Similar outcomes across varying empiric treatment regimens suggest that there is an opportunity to use narrow-spectrum antibiotic therapies and limit the use of antibiotics with antipseudomonal activity. This process of decreasing use of inappropriate broad-spectrum antibiotics is consistent with the CDC’s core elements for antibiotic stewardship [28] and could potentially help to decrease the incidence of developing CDI and subsequent antibiotic-resistant infections. As with all studies of this nature, the findings need to be confirmed in prospective studies. For now, these findings can serve as the basis for an antimicrobial stewardship initiative to assess the use of broad-spectrum antibiotics among lower-risk patients with CA cIAI.

Classification and Corresponding ICD-10 Diagnosis Codes [2]
Sepsis, severe sepsis, septic shock
A40 Streptococcal sepsis A41 Other sepsis A48.3 Toxic shock syndrome T81.12 Postprocedural septic shock R65.2 Severe sepsis R65.20 Severe sepsis without septic shock R65.21 Severe sepsis with septic shock R65.X SIRS, unspecified
Components of sepsis
Fever R50 Fever of other and unknown origins Tachycardia R00.0 Tachycardia, unspecified R00.2 Palpitations, tachypnea R06.82 Tachypnea, unspecified R06.02 Shortness of breath, altered mental status R41.82 Altered mental status, unspecified; hypoglycemia E16.1 Other hypoglycemia E16.2 Hypoglycemia, unspecified; hyperglycemia R73.9 Hyperglycemia, unspecified; leukocytosis
D72.82 Elevated white blood cell count D72.825 Bandemia D72.828 Other elevated white blood cell count D72.829 Elevated white blood cell count, unspecified; leukopenia D72.81 Decreased white blood cell count; hypotension I95.X Hypotension R03.1 Nonspecific low blood pressure reading, hypoxemia R06.00 Dyspnea, unspecified R09.02 Hypoxemia, oliguria R34.X Oliguria N18.6 End-stage renal disease N17.X Acute kidney failure N19 Unspecified kidney failure N99.0 Postprocedural (acute) (chronic) kidney failure, coagulation abnormalities D65 Disseminated intravascular coagulation [defibrination syndrome] R79.1 Abnormal coagulation profile, thrombocytopenia D69.4 Other primary thrombocytopenia D69.5 Secondary thrombocytopenia D69.6 Thrombocytopenia, unspecified; hyperbilirubinemia E80.7 Disorder of bilirubin metabolism, unspecified; acute lung injury/acute respiratory failure; respiratory failure J96.X Respiratory failure, not elsewhere classified J95.82 Postprocedural respiratory failure J96 Respiratory failure, not elsewhere classified J98.1 Pulmonary collapse J96.0 Acute respiratory failure J96.00 Acute respiratory failure, unspecified whether with hypoxia or hypercapnia
J96.01 Acute respiratory failure with hypoxia J96.02 Acute respiratory failure with hypercapnia Z99.11 Dependence on respirator (ventilator) J96.X Respiratory failure, not elsewhere classified
Physiologic risk factors
Advanced age ≥70 y
Malignancy C00, C01, C02, C03, C04, C05, C06, C07, C08, C09, C10, C11, C12, C13, C14, C15, C16, C17, C18, C19, C20, C21, C22, C23, C24, C25, C26, C30, C31, C32, C33, C34, C37, C38, C39, C40, C41, C43, C45, C46, C47, C48, C49, C50, C51, C52, C53, C54, C55, C56, C57, C58, C60, C61, C62, C63, C64, C65, C66, C67, C68, C69, C70, C71, C72, C73, C74, C75, C76, C81, C82, C83, C84, C85, C88, C90, C91, C92, C93, C94, C95, C96, C97
Kidney dysfunction or significant renal disease N18.6 End-stage renal disease N17.X Acute kidney failure N19 Unspecified kidney failure I12.0 Hypertensive chronic kidney disease with stage 5 chronic kidney disease or end-stage renal disease I13.11 Hypertensive heart and chronic kidney disease without heart failure with stage 5 chronic kidney disease or end-stage renal disease N99.0 Postprocedural (acute) (chronic) kidney failure I13.2 Hypertensive heart and chronic kidney disease with heart failure and with stage 5 chronic kidney disease or end-stage renal disease N18.6 End-stage renal disease
Hepatic dysfunction/significant liver disease or cirrhosis K91.82 Postprocedural hepatic failure K70 Alcoholic liver disease K71 Toxic liver disease K72 Hepatic failure, not elsewhere classified K74 Fibrosis and cirrhosis of liver K75 Other inflammatory liver diseases K76 Other diseases of liver K77 Liver disorders in diseases classified elsewhere B15-B19 hepatitis P78.81 congenital cirrhosis of liver
Hypoalbuminemia R77.0 Abnormality of albumin
Significant cardiovascular compromise R94.3 Abnormal results of CV function studies R55 Syncope and collapse

Derived from Mazuski et al. [2].

Abbreviations: CV, Cardiovascular; ICD-10, International Classification of Diseases, Tenth Revision; SIRS, Systemic inflammatory response syndrome.