Age and Its Impact on Outcomes with Intra-Abdominal Infection

Abstract

Background:

Age has been shown to play a significant role in the etiology of complicated intra-abdominal infections (cIAIs), but the correlation between age and outcomes after therapy was not investigated in the Study to Optimize Peritoneal Infection Therapy (STOP-IT) trial.

Patients and Methods:

Data were obtained by post hoc analysis of the STOP-IT trial database. Patients were stratified by age <65 or ≥65 years. Primary outcomes were surgical site infection (SSI), recurrent IAI (recIAI), and death. Multivariable analysis was performed to identify independent predictors of outcomes.

Results:

There were 398 subjects <65 and 120 ≥ 65 years. Overall baseline characteristics of the two groups were similar. The site of infection was similar between groups except: Colon or rectum (48.3% vs. 29.9%, p = 0.0002) and biliary tree (16.7% vs. 9.1%, p = 0.02), which were more common in the older group, whereas small intestine (6.7% vs. 16.3%, p = 0.008) and appendix (4.2% vs.17.1%, p = 0.0004) were more common in the younger group. Among the primary outcomes, only death was significantly different between the age groups and was more prevalent in the ≥65 years group (4 [3.3%] vs. 1 [0.3%], p = 0.01). Surgical site infection (9.2% vs. 7.3%, p = 0.50), recIAI (15.8% vs. 14.4%, p = 0.69), and a composite outcome (26.7% vs. 20.4%, p = 0.14) were statistically similar between the age groups, and this remained true when controlling for other co-variables. Multivariable analyses did not reveal age as an independent predictor of the composite or individual outcomes.

Conclusion:

Patients with a more advanced age demonstrated variable sources of infection relative to the younger cohort, yet received similar treatments. Patient age was not an independent predictor of the undesired cIAI outcomes. These findings suggest that advanced age itself does not play a significant role in predicting these adverse outcomes for cIAIs and does not necessitate an altered treatment tactic.

Centers for Medicare and Medicaid along with the Agency for Healthcare Research and Quality have defined quality metrics that are measured and tracked allowing for comparisons between institutions. Patient safety indicators [1], which include post-operative sepsis, wound complications, and death, are included as focus items, the rates of which are used to determine reimbursements to hospitals. As a result, the healthcare industry is placing more emphasis on understanding the causes of complications and opportunities to optimize outcome.

The Study to Optimize Peritoneal Infection Therapy (STOP-IT) is one example of an attempt to delineate best therapy in patients with intra-abdominal infections [2]. In this randomized trial, patients with short-course antibiotic therapy were found to have similar outcomes defined as surgical site infections (SSI), recurrent intra-abdominal infection (recIAI), and death as those who were treated longer. Although this trial certainly offers support to the practice of good antimicrobial stewardship, concerns may remain as to whether these findings are applicable to all patient populations.

One population that merits unique consideration is the elderly. The number of patients over the age of 65 years continues to grow. Average life expectancy in the United States has increased from 70.2 years in 1965 to a predicted 80 years in 2020 [3]. As a result, more elderly patients have intra-abdominal pathology and infections with and without surgical procedures. In fact, age is an independent predictor for some of these complications including SSIs [4]. Given this fact, clinicians may employ treatment algorithms in the elderly population, which may not be appropriate. The purpose of this study was to determine whether patients 65 years and older with complicated IAIs (cIAIs) have different outcomes compared with their younger counterparts when treated with similar regimens.

Patients and Methods

Patient population and data source

Patient data were collected from the database used in the Study to Optimize Peritoneal Infection Therapy (STOP-IT) trial to perform post hoc analysis to examine the effect of age on post-treatment outcomes of cIAIs. The STOP-IT trial was an open-label, multi-center, randomized controlled trial conducted to determine the effect of antimicrobial duration on treatment outcomes after adequate source control of cIAIs. The original STOP-IT publication contains the detailed study design description with detailed inclusion and exclusion criteria [2]. Subjects were categorized into two groups based on age: Less than or greater than or equal to 65 years.

Data collection and outcomes

Characteristics of the groups including patient demographics, co-morbidities, clinical indicators of disease severity, setting of infection acquisition, and site of the infection were collected and stratified by age. To determine whether differences existed in treatment modalities based on site of infection and age, further categorization of source-control procedures and antimicrobial treatment regimens was conducted based on source of infection (i.e., sites with statistical differences between age groups) and stratified by age.

Antimicrobial agents were included only if recorded as administered at least once during treatment course and provided to a minimum of 25 patients (∼5% of cohort). Additional grouping of antimicrobial agents included anti-pseudomonal carbapenems (imipenum-cilastatin and meropenem), cephamycins (cefoxitin and cefotetan), and fluoroquinolones (cipro-, levo-, and moxifloxacin). The SSI, recIAI, death, and a composite metric of all three outcomes constituted the undesirable outcomes measured. Measurement of these complications in the original trial occurred at 30 days [2].

Statistical analysis

Descriptive statistics were analyzed and reported for baseline characteristics as mean ± standard deviation were applicable. Uni-variable analyses were performed to determine differences in collected variables based on age. Categoric variables were compared using either Pearson chi-square testing or, in cases of small cell sizes, the Fisher exact test. Continuous variables were compared using two-tailed, independent samples t-test. Effects of age on outcomes were analyzed by uni-variable and multivariable logistic regression models. Each multivariable model was characterized by outcome metrics as the dependent variable and age as an independent variable. The coefficient for age in a model was considered a measure of its effect on the outcome. Data were analyzed using SAS 9.3 for Windows (SAS Institute, Cary, NC).

Results

From the original 518 patients included in the STOP-IT trial, stratification by age resulted in 120 subjects 65 years or greater and 398 subjects less than 65 years. Table 1 shows baseline characteristics stratified by age. A similar distribution in gender, race, and co-morbid conditions at baseline was observed for the two age groups. On average, however, the older cohort had a lower baseline body mass index (27.7 ± 7.0 vs. 29.4 ± 9.3, p = 0.03) and a higher Acute Physiology and Chronic Health Evaluation (APACHE II) score (13.4 ± 5.4 vs. 9.0 ± 5.8, p < 0.0001). When the age component of APACHE was removed from the calculation, the difference between the groups disappeared (8.1 ± 5.4 vs. 7.6 ± 5.5, p = 0.42).

Table 1.

Baseline Characteristics Stratified by Age

Characteristic	Age <65 years (n = 398)	Age ≥65 years (n = 120)	p^a
Age, mean (SD, min, max)	45.8 (12.2,16,64)	73.4 (6.1,65,88)	0.0001
Gender, no. (%)
Male	217 (54.5)	72 (60.0)	0.30
Race, no. (%)			0.06
White	301 (75.6)	103 (85.8)
Black	80 (20.1)	14 (11.7)
Other^b	17 (4.3)	3 (2.5)
Hispanic, no. (%)	29 (7.3)	6 (5.0)	0.38
Body mass index, mean ± SD	29.4 ± 9.3	27.7 ± 7.0	0.03
Co-morbid conditions, no. (%)
Diabetes mellitus	60 (15.1)	18 (15.0)	0.98
Steroid use	15 (3.8)	1 (0.8)	0.14
Malignant disease	40 (10.1)	19 (15.8)	0.08
Chronic kidney disease	15 (3.8)	1 (0.8)	0.14
Dialysis dependence	7 (1.8)	2 (1.7)	1.00
Chronic liver disease	15 (3.8)	3 (2.5)	0.78
Chronic pulmonary disease	22 (5.5)	6 (5.0)	0.82
Red cell transfusion	33 (8.3)	10 (8.3)	0.99
Cardiac disease	49 (12.3)	21 (16.4)	0.15
Inflammation of bowel	45 (11.3)	8 (6.7)	0.14
Lung disease	20 (5.0)	4 (3.3)	0.62
Vascular disease	23 (5.8)	3 (2.5)	0.23
APACHE II score, mean ± SD	9.0 ± 5.8	13.4 ± 5.4	<0.0001

SD = standard deviation; APACHE = Acute Physiology and Chronic Health Evaluation.

t-test used for body mass index; ^bAmerican Indian/Alaskan, Asia, or Unknown.

Chi-square used for all other variables except where cell <5 in which the Fisher exact test was used.

Table 2 represents infection characteristics and treatment modalities used stratified by age. The majority of subjects acquired the index infection in the community setting with an overall even distribution between age groups. Older patients had a higher prevalence of the colon or rectum (48.3% vs. 29.9%, p = 0.0002) and biliary tree (16.7% vs. 9.1%, p = 0.02) constituting the site of infection, whereas younger patients had higher rates of the small intestine (6.7% vs. 16.3%, p = 0.008) and appendix (4.1% vs. 17.1%, p = 0.0004) constituting the site of infection for the cIAI. Source control procedures used were not significantly different between age groups. Antimicrobial agent utilization remained balanced between the groups except for ertapenem and ceftriaxone, which had increased use in younger patients, 4.2% vs. 12.1%, p = 0.01 and 0.8% vs. 6.5%, p = 0.01, respectively. Total antibiotic days were not different between age categories.

Table 2.

Infection Characteristics and Treatment Strategies Stratified by Age, n = 518

Characteristic	Age <65 (n = 398)	Age ≥65 (n = 120)	p
Maximum white blood cell count, per mm³, mean ± SD	16.7 ± 10.1	15.3 ± 7.5	0.09
Maximum body temperature, °C, mean ± SD	37.8 ± 0.9	37.5 ± 0.8	0.002
Setting of acquisition of cIAI, no. (%)			0.19
Community-acquired	240 (60.5)	81 (67.5)
Healthcare-associated	105 (26.5)	22 (18.3)
Hospital-acquired	52 (13.1)	17 (14.2)
Site of Infection, no. (%)
Colon or rectum	119 (29.9)	58 (48.3)	0.0002
Small intestine	65 (16.3)	8 (6.7)	0.008
Appendix	68 (17.1)	5 (4.2)	0.0004
Biliary tree (including gallbladder)	36 (9.1)	20 (16.7)	0.02
Duodenum	16 (4.0)	7 (5.8)	0.40
Esophagus	3 (0.8)	0 (0.0)	1.00
Stomach	24 (6.0)	7 (5.8)	0.94
Pancreas	13 (3.3)	3 (2.5)	1.00
Liver	15(3.8)	3 (2.5)	0.78
Abdominal wall surgical site	8 (2.0)	5 (4.2)	0.19
Source-control procedure, no. (%)
Percutaneous drainage	135 (33.9)	37 (30.8)	0.53
Resection and anastomosis or closure	97 (24.4)	35 (29.2)	0.29
Surgical drainage only	86 (21.6)	23 (19.2)	0.57
Resection and proximal diversion	48 (12.1)	16 (13.3)	0.71
Simple closure	24 (6.0)	8 (6.7)	0.80
Surgical drainage and diversion	8 (2.0)	1 (0.8)	0.69
Antimicrobial agents, no. (%)
Amoxicillin or amoxicillin-clavulanate	38 (9.6)	12 (10.0)	0.88
Piperacillin-tazobactam	208 (52.3)	75 (62.5)	0.05
Carbapenem
Ertapenem	48 (12.1)	5 (4.2)	0.01
Meropenem/imipenum-cilastatin	20 (5.0)	5 (4.2)	0.70
Ceftriaxone	26 (6.5)	1 (0.8)	0.01
Cephamycins^a	28 (7.0)	6 (5.0)	0.43
Fluoroquinolone^b	108 (27.1)	33 (27.5)	0.94
Metronidazole	128 (32.2)	35 (29.2)	0.54
Vancomycin	94 (23.6)	37 (30.8)	0.11
Total antibiotic days, mean ± SD	7.4 ± 5.7	7.5 ± 4.9	0.95

SD = standard deviation; cIAI = complicated intra-abdominal infection.

cefotetan and cefoxitin; ^bciprofloxacin, levofloxacin, and moxifloxacin.

Statistical differences in the site of infection distribution by age stratification prompted a more detailed sub-group analysis by the site of infection and either the source-control procedure or antimicrobial used (Tables 3 and 4). No significant differences existed in the source-control procedure used across the sites of infection included regardless of age categorization. Antimicrobial agents used were relatively similar between the two age groups relative to site of infection.

Table 3.

Frequencies of Source-Control Procedures by Age and Site of Infection

	Site of infection
	Colon or rectum			Small intestine			Appendix			Biliary tree
	n = 177			n = 73			n = 73			n = 56
	<65	≥65		<65	≥65		<65	≥65		<65	≥65
Source-control Procedure	n = 119	n = 58	p^a	n = 65	n = 8	p^a	n = 68	n = 5	p^a	n = 36	n = 20	p^a
Percutaneous drainage, no. (%)	38 (31.9)	12 (20.7)	0.15	20 (30.8)	3 (37.5)	0.70	26 (38.2)	0 (0.0)	0.15	10 (27.8)	7 (35.0)	0.76
Resection and anastomosis or closure, no. (%)	29 (24.4)	18 (31.0)	0.37	17 (26.2)	2 (25.0)	1.00	19 (27.9)	1 (20.0)	1.00	12 (33.3)	6 (30.0)	1.00
Surgical drainage only, no. (%)	28 (23.5)	12 (20.7)	0.71	12 (18.5)	2 (25.0)	0.64	14 (20.6)	2 (40.0)	0.30	9 (25.0)	4 (20.0)	0.75
Resection and proximal diversion, no. (%)	17 (14.3)	11 (19.0)	0.51	11 (16.9)	1 (12.5)	1.00	6 (8.8)	2 (40.0)	0.09	2 (5.6)	0 (0.0)	0.53
Simple closure, no. (%)	4 (3.4)	5 (8.6)	0.16	4 (6.2)	0 (0.0)	1.00	2 (2.9)	0 (0.0)	1.00	2 (5.6)	2 (10.0)	0.61
Surgical drainage and diversion, no. (%)	3 (2.5)	0 (0.0)	0.55	1 (1.5)	0 (0.0)	1.00	1 (1.47)	0 (0.0)	1.00	1 (2.8)	1 (5.0)	1.00

Fisher exact test.

Table 4.

Frequencies of Selected Antimicrobial Treatment Stratified by Age and Site of Infection

	Site of infection
	Colon or rectum			Small intestine			Appendix			Biliary tree
	n = 177			n = 73			n = 73			n = 56
	<65	≥65		<65	≥65		<65	≥65		<65	≥65
Antimicrobial	n = 119	n = 58	p^a	n = 65	n = 8	p^a	n = 68	n = 5	p^a	n = 36	n = 20	p^a
Amoxcillin or Amoxicillin-clavulanate, no. (%)	6 (5.0)	2 (3.5)	1.00	4 (6.2)	1 (12.5)	0.45	9 (13.2)	0 (0.0)	1.00	6 (16.7)	5 (25.0)	0.50
Piperacillin-tazobactam, no. (%)	61 (51.3)	36 (62.1)	0.27	30 (46.2)	4 (50.0)	1.00	21 (30.9)	4 (80.0)	0.04	28 (77.8)	13 (65.0)	0.35
Ertapenem, no. (%)	17 (14.3)	1 (1.7)	0.01	7 (10.8)	2 (25)	0.25	11 (16.2)	1 (20.0)	1.00	1 (2.8)	1 (5.0)	1.00
Meropenem/ imipenum-cilastatin, no. (%)	6 (5.0)	3 (5.2)	1.00	4 (6.2)	0 (0.0)	1.00	1 (1.5)	0 (0.0)	1.00	3 (8.3)	1 (5.0)	1.00
Ceftriaxone, no. (%)	6 (5.0)	0 (0.0)	0.18	7 (10.8)	0 (0.0)	1.00	9 (13.2)	0 (0.0)	1.00	0 (0.0)	1 (5.0)	0.36
Cephamycin,^b no. (%)	9 (7.6)	1 (1.7)	0.17	5 (7.7)	1 (12.5)	0.51	7 (10.3)	0 (0.0)	1.00	3 (8.3)	0 (0.0)	0.55
Fluoroquinolone,^c no. (%)	37 (31.1)	19 (32.8)	0.86	18 (27.7)	1 (12.5)	0.67	23 (33.8)	2 (40.0)	1.00	10 (27.8)	5 (25.0)	1.00
Metronidazole, no. (%)	39 (32.8)	18 (31.0)	0.87	21 (32.3)	1 (12.5)	0.42	32 (47.1)	2 (40.0)	1.00	6 (16.7)	8 (40.0)	0.10
Vancomycin, no. (%)	35 (29.4)	18 (31.0)	0.86	12 (18.5)	3 (37.5)	0.35	6 (8.8)	0 (0.0)	1.00	8 (22.2)	6 (30.0)	0.54

Fisher exact test; ^bcefotetan and cefoxitin; ^cciprofloxacin, levofloxacin, and moxifloxacin.

Notable differences existed for two sites of infection regarding the antimicrobial agents selected. Ertapenem was more often used for colon and rectal sites for younger patients (1.7% vs. 14.3%, p = 0.01); however, piperacillin-tazobactam utilization was favored for older patients with the appendix documented as the site of infection (80.0% vs. 30.9%, p = 0.04).

Uni-variable analysis revealed no difference in the composite outcome with 26.7% of subjects age ≥65 years and 20.4% of subjects <65 years demonstrating an undesirable outcome (p = 0.15). Individual components except death, which was more prevalent in the older group (3.3% vs. 0.3%, p = 0.01), were not statistically different between age groups: SSI (9.2% vs. 7.3%, p = 0.50) and recIAI (15.8% vs. 14.4%, p = 0.69). Multivariable analyses demonstrated no differences in the 30-day composite outcome or individual outcomes when stratified by age.

Discussion

As the population of the United States continues to age, it has become increasingly important to determine which factors are critical in the care of the elderly patient with an IAI. More than 23% of the patients in the recently published STOP-IT trial were age 65 years or older emphasizing the importance of understanding optimal therapy for patients in this population. Early source control and antimicrobial therapy comprise the mainstay treatment for all patients with IAIs, but STOP-IT demonstrated that increasing duration of antibiotic treatment beyond four days was not beneficial for patients with IAIs. Using data collected from this study, it was our aim to examine how outcomes of IAIs among the elderly compared with their younger counterparts and identify differences to provide insight into opportunities for improvement.

Advanced age has been suggested as an independent risk factor for death in patients with cIAIs [5,8]. Further, guidelines for management of cIAIs published by the Surgical Infection Society (SIS) and Infectious Diseases Society of America (IDSA) list advanced age as a factor for predicting failure of source control. These guidelines further stratify patients with advanced age into the high-risk category for both biliary and complicated extra-biliary IAI, and recommend escalated antimicrobial therapy for empiric treatment of patients with community-acquired infections in this population [6]. Moreover, patients with advanced age are also particularly vulnerable to the virulent effects of certain pathogens including enterococci, which have been shown to have a >50% mortality rate when associated with peritoneal infections in patients 75 years and older [7]. Results from the current study contradict previous reports that suggest more treatment in the elderly.

The composite outcome of death, SSI, and recIAI was similar between the groups. These results support the original trial that there was no correlation between composite outcome and age as a continuous variable (odds ratio) 0.99 (0.98–1.01), p = 0.29) [2]. Death was increased in the over 65 years population, however. Because of its rare occurrence (n < 5 in each group), its true clinical significance is questionable, particularly because it is a global outcome measure and may not have resulted specifically from the IAI.

We have shown that there were no significant differences in outcomes in SSIs and recIAIs in patients older than 65 years. These findings were confirmed by both uni-variable and multivariable analysis that standardized the groups. These outcomes were achieved with both similar overall approaches to source control and duration of therapy between the groups demonstrating that a similar approach regardless of age resulted in similar overall outcomes.

These findings are important because they call into question the recommendations published by the SIS and IDSA suggesting a more aggressive approach, including broader spectrum antimicrobial coverage. These guidelines based a heightened severity category with subsequent escalation in therapy on the ability of age (in particular age >70 years) to predict poor source control and death [7]. Of note, the studies demonstrating a link between age and undesirable outcomes dealt primarily with more severe manifestations of cIAIs, notably post-operative infections, and carried higher overall death relative to the patients in the present study [8,9]. By contrast, another larger study containing a more diverse set of cIAI presentations demonstrated that age was not an independent predictor of post-operative death [10].

More contemporary data that includes a population more reflective of the present study, has confirmed a role for age in predicting death, but in tandem with other factors [5,11). These findings suggest the need for a defined multi-factorial risk factor system for predicting severity and death, thus calling into question the utility of age as a stand-alone predictor of disease severity. In addition, age did not predict the isolation of resistant pathogens, even in a cIAI population with a majority of healthcare-associated infections [12], indicating that augmented antimicrobial coverage based on age alone requires further investigation.

It is important to recognize that some differences did exist between the groups that prompted further investigation to ensure that these did not skew the results. The APACHE scores were higher in the more senior group, but this is of little value considering age is a component of the APACHE scoring system. Further, when we removed age from the calculations, the groups were similar. The finding of potential clinical value was that of the distribution of infection location. The colorectal sources accounted for a larger percentage of IAI in the older group as did the biliary tree/gallbladder. Sources of IAI more common in the <65 population were the appendix and the small intestine. These outcomes are consistent with the current knowledge of diverticular disease, which increases in prevalence with advancing age [8,13]. Conversely, appendicitis tends to be a disease of the young with an estimated incidence rate of only 1 in 2000 persons after the age of 65.

Although no statistically significant differences in source control procedure were noted based on site of infection, percutaneous drainage was more commonplace among the under the age of 65 group for both colorectal and appendiceal sources. Resection and diversion was more common among the older group for infections resulting from appendicitis. It has been demonstrated that patients 65 years and older present later in the course of their disease and are several times more likely than their younger counterparts to present with a generalized peritonitis rather than a localized, drainable abscess [9,14]. In our trial, however, these cases are so few that determination of the true implication is difficult to determine.

Likewise, the antimicrobial therapy administered was statistically similar between the study groups except in two categories. For IAIs of colorectal etiology, ertapenem was used more frequently in patients under the age of 65. For patients with appendicitis, piperacillin-tazobactam was prescribed with more regularity to patients 65 or older. The SIS and IDSA guidelines for initial empiric therapy for complicated IAIs recommend escalation of antimicrobial therapy in patients of advanced age. These findings suggest that adherence to these guidelines may be inconsistent, yet differences in outcomes between the two groups were not statistically significant.

Limitations

Although this database was excellent for examining cIAIs as a whole and had detailed records capable of providing information for more specific questions, the volume of patients was not adequate to provide the statistical power needed to examine differences among small sub-sets of the population. Many data points of interest had fewer than five occurrences, which puts our findings at risk for a type II error. Moreover, although 23% of the patients were elderly, there is the potential for a selection bias in that sicker elderly patients may not have been included. Such a bias could impede our ability to fully understand the pathophysiology of this disease process in this age group.

Conclusions

The goal of management of IAI is eradication of infection without recurrence. Despite distinct differences in etiology, this study confirms that a similar approach is important such that all patients should be aggressively treated with source control and antibiotics. It also demonstrates, however, that short course therapy can be applied to the elderly population emphasizing the need to re-evaluate guidelines that recommend a different approach to these patients. Further study into the role of age and its impact on these outcomes is warranted.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

AHRQ Quality Indicators: Patient safety indicators. qualityindicators.ahrq.gov/Downloads/Modules/PSI/V50/PSI_Brochure.pdf (Last accessed November 11, 2016).

Sawyer

, Claridge

, Nathens

, et al. Trial of short-course antimicrobial therapy for intraabdominal infection. N Engl J Med, 2015; 372:1996–2005.

Data360.org. Life Expectancy–United States. 2011. www.data360.org/dsg.aspx?Data_Set_Group_Id=195 (Last accessed March 11, 2016).

Neumayer

, Hosokawa

, Itani

, et al., Multivariable predictors of postoperative surgical site infection after general and vascular surgery: Results from the patient safety in surgery study. J Am Coll Surg, 2007; 204:1178–1187.

Inui

, Haridas

, Claridge

, Malangoni

. Mortality for intra-abdominal infection is associated with intrinsic risk factors rather than the source of infection. Surgery, 2009; 146:654–661.

Solomkin

, Mazuski

, Bradley

, et al. Diagnosis and management of complicated intra-abdominal infection in adults and children: Guidelines by the Surgical Infection Society and the Infectious Diseases Society of America. Clin Infect Dis, 2010; 50:133–164.

Dupont

, Friggeri

, Touzeau

, et al. Enterococci increase the morbidity and mortality associated with severe intra-abdominal infections in elderly patients hospitalized in the intensive care unit. J Antimicrob Chemother, 2011; 66:2379–2385.

Koperna

, Schulz

. Relaparotomy in peritonitis: Prognosis and treatment of patients with persisting intraabdominal infection. World J Surg, 2000; 24:32–37.

Mulier

, Penninckx

, Verwaest

, et al. Factors affecting mortality in generalized postoperative peritonitis: Multivariate analysis in 96 patients. World J Surg, 2003; 27:379–384.

10.

Pacelli

, Doglietto

, Alfieri

, et al. Prognosis in intra-abdominal infections. Multivariate analysis on 604 patients. Arch Surg, 1996; 131:641–645.

11.

Sartelli

, Catena

, Ansaloni

, et al. Complicated intra-abdominal infections in Europe: A comprehensive review of the CIAO study. World J Emerg Surg, 2012; 7:36.

12.

Swenson

, Metzger

, Hedrick

, et al. Choosing antibiotics for intra-abdominal infections: what do we mean by “high risk”?. Surg Infect (Larchmt), 2009; 10:29–39.

13.

Humes

, Simpson

, Spiller

. Colonic diverticular disease. BMJ Clin Evid, 2007; 2007:0405.

14.

Watters

, Blakslee

, March

, Redmond

. The influence of age on the severity of peritonitis. Can J Surg, 1996; 39:142–146.