Biomarkers (Procalcitonin,C Reactive Protein,and Lactate) as Predictors of Mortality in Surgical Patients with Complicated Intra-Abdominal Infection

Abstract

Background:

An accurate and readily available biomarker for identifying patients with complicated intra-abdominal infection needing special attention in critical care units because of their greater risk of dying would be of value for intensivists.

Methods:

A multi-center, observational, retrospective study explored blood lactate, C-reactive protein (CRP), and procalcitonin (PCT) concentrations, and also Sequential Organ Failure Assessment (SOFA) and Simplified Acute Physiology Score (SAPS II) as mortality predictors in all adult patients with complicated intra-abdominal infection (cIAI) admitted to Surgical Critical Care Units (SCCUs) for ≥48 h in four Spanish hospitals (June 2012–June 2013). Logistic regression models (step-wise procedure) were constructed using as dependent variables “intra-SCCU mortality” or “overall mortality,” and variables showing differences (p≤0.1) in bivariate analyses as independent variables.

Results:

One hundred twenty-one cases were included. Mortality intra-SCCU (R²=0.189, p=0.001) was associated with SAPS II (categorized as high if ≥47) (OR=9.55; 95% CI, 1.09–83.85; p=0.042) and 24 h-lactate (≥5.87 categorized as high) (OR=6.90; 95% CI, 1.28–37.08). Overall mortality (R²=0.275, p=0.001) was associated with peak PCT (≥100 categorized as high) (OR=11.28; 95% CI, 1.80–70.20), peak lactate (≥1.8 categorized as high) (OR=8.86; 95% CI, 1.51–52.10) and SOFA at admission (≥7 categorized as high) (OR=8.14; 95% CI, 1.69–39.20), but was predicted better (R²=0.275, p=0.001) by a single dummy variable (high peak PCT-high peak lactate concentrations) (OR=99.11; 95% CI, 5.21–1885.97; p=0.002).

Conclusions:

In the present study, SAPS II and 24 h-lactate concentrations predicted intra-SCCU mortality whereas overall mortality was predicted better by concurrent high PCT and lactate peak concentrations than by clinical scores or by each biomarker separately.

An assessment tool to identify surgical critical care unit (SCCU) patients with complicated intra-abdominal infection who are at greater risk of fatal outcome would be of value to ensure the optimization of healthcare resources. Clinical scoring systems allow satisfactory prediction of overall prognosis, the Sequential Organ Failure Assessment (SOFA) [1] and the Simplified Acute Physiology Score (SAPS II) [2] being well-validated scores for risk assessment. However, scoring systems for decision making in patients with sepsis have been criticized [3].

An alternative or complementary approach is the use of blood biomarkers mirroring host response to infection and, indirectly, severity of infection. These biochemical parameters for stratifying severity have been studied poorly in intra-abdominal infections, with scant published studies [4 –6], inclusion of non-comparable study populations, and use of non-standardized assay techniques [4]. Among the common biomarkers used, C-reactive protein (CRP) is a useful marker of systemic inflammation [7] without discriminating infection from other inflammatory processes. Elevated serum lactate concentrations reflect disrupted protein metabolism and are believed to be marker of tissue hypoxia [8]. Lastly, procalcitonin (PCT) concentrations increase during bacterial [9] but not viral infections or non-infectious inflammatory reactions [10], and remain unaffected by the administration of corticosteroids when compared with other biomarkers such as CRP [11].

An accurate and readily available biomarker for identifying critical care patients with complicated intra-abdominal infection needing special attention because of their greater risk of dying would be of value for intensivists. The aim of this study was to explore the utility of blood lactate concentrations, CRP and PCT, and additionally SOFA and SAPS II scores, as predictors of mortality in patients with complicated intra-abdominal infection admitted to the SCCU.

Patients and Methods

A multi-center observational study was performed in four Spanish hospitals from June 2012 to June 2013. Retrospective analysis was performed on prospectively acquired data recorded as part of daily routine care in medical records of all adult patients with complicated intra-abdominal infection requiring surgery and SCCU admission for ≥48h. Ethical approval for this study was provided by the Ethical Committee of Hospital Universitario La Paz, Madrid, Spain. Informed consent was waived because of the observational nature of the study.

The SAPS II and SOFA scores were calculated with data in the first 24 h and also with data recorded 72 h after admission in the case of the SOFA score. Values of CRP, lactate concentrations and PCT concentrations (ThermoFisher Scientific Inc., Madrid, Spain) determined 24 h, 48 h, and 72 h after admission were recorded. For each patient, the highest value of each marker was considered as peak value.

Mortality was assessed as intra-SCCU mortality and overall mortality. Intra-SCCU mortality was defined as mortality during stay in the SCCU and overall mortality was the mortality occurring in the period of 28 d from SCCU admission, and included the intra-SCCU mortality.

Comparisons between proportions were performed by the Chi-square test and the Fisher exact test, when necessary. For quantitative variables, because data did not show normality in the Kolmogorov-Smirnoff test, the Kruskal-Wallis, and Mann-Whitney tests when necessary, were used. All variables were compared between survivors and non-survivors. The discriminatory power of variables showing significance in the bivariate analysis was evaluated by the area under the receiver-operating characteristics (ROC) curve (AUC) and the 95% confidence interval (95% CI). Logistic regression models (step-wise procedure) were performed using as dependent variables “intra-SCCU mortality” or “overall mortality” and as independent variables those showing differences (p≤0.1) in bivariate analyses. Interactions and linear dependence between independent variables were previously controlled. In addition, biomarkers significant in the multivariable analysis were used to create dummy variables that, if significant (p<0.05), were also included in the model. The models showing the maximum parsimony (the lowest number of variables with no significant reduction in the value of the determination coefficient) and the highest R² were considered. Statistical analyses were performed using SPSS version 14 (IBM Inc., Armonk, NY).

Results

A total of 121 cases were included in the study; mean age of patients was 65.6±16.9 y (range 18–96 y), and 59.5% were males. Patients were admitted in the SCCU after surgery involving the following surgical sites: Colon (50.8%), small bowel (18.5%), appendix (10.5%), biliary tract (8.9%), gastroduodenal (8.1%), and others (3.2%). Mortality was 11.6% (14 of 121 patients) during the SCCU stay and overall mortality was 18.2% (22 of 121).

Tables 1 and 2 show values of clinical scores and biomarkers according to intra-SCCU and overall mortality, respectively. Values of the two clinical scores studied were greater in those patients that finally died than in those surviving, both considering intra-SCCU and overall mortality. With respect to biomarkers, no differences in CRP values were found by outcome. By contrast, 24 h-, 48 h- and peak values of lactate, and 24 h-, 48 h-, 72 h- and peak values of PCT were greater in patients dying (both for intra-SCCU mortality and overall mortality).

Table 1.

Clinical Scores and Biomarkers in Relation to Intra-SCCU Mortality Median: (Interquartile Range) Except Where Indicated

	Survival (n=107)	Non survival (n=14)	p
Clinical scores
SAPS II
0 h	42.00 (30.00, 51.00)	64.50 (50.75, 70.50)	<0.001
0 h ≥47 [n (%)]	41 (38)	12 (86)	0.001
SOFA
0 h	5.00 (3.00, 9.00)	9.00 (8.00, 10.25)	<0.001
0 h ≥8 [n (%)]	34 (32)	12 (86)	<0.001
72 h	3.00 (2.00, 6.25)	8.00 (4.50, 11.50)	0.001
Biomarkers
CRP (mg/L)
24 h	208.00 (144.70, 280.00)	246.00 (184.75, 137.50)	0.222
48 h	230.00 (163.00, 305.20)	209.00 (139.00, 317.00)	0.663
72 h	170.55 (123.00, 257.50)	156.00 (84.40, 251.50)	0.769
Peak	264.30 (189.23, 337.28)	287.10 (226.18, 366.00)	0.524
Lactate (mmol/L)
24 h	1.40 (1.00, 2.00)	2.60 (1.85, 4.90)	<0.001
24 h ≥1.90 [n (%)]	25 (28.4)	10 (76.9)	0.001
24 h ≥5.87 [n (%)]	2 (2.3)	3 (23.1)	0.015
48 h	1.10 (0.90, 1.53)	2.10 (1.15, 3.13)	0.006
72 h	1.10 (0.80, 1.50)	1.20 (1.00, 2.35)	0.188
Peak	1.90 (1.20, 2.95)	3.60 (2.18, 5.63)	0.001
PCT (ng/mL)
24 h	7.75 (3.49, 26.83)	38.00 (7.74, 121.06)	0.034
24 h ≥5.80 [n (%)]	55 (60.4)	10 (83.3)	0.202
48 h	6.32 (3.27, 21.92)	25.48 (9.90, 58.39)	0.019
72 h	4.68 (1.92, 20.23)	17.97 (10.38, 31.75)	0.014
Peak	8.81 (3.45, 28.38)	38.00 (14.64, 131.93)	0.006

SCCU=Surgical Critical Care Units; SAPS=Simplified Acute Physiology Score; SOFA=Sequential Organ Failure Assessment; CRP=C-reactive protein; PCT=procalcitonin.

Table 2.

Clinical Scores and Biomarkers in Relation to Overall Mortality: Median (Interquartile Range) Except Where Indicated

	Survival (n=99)	Non survival (n=22)	p
Clinical scores
SAPS II
0 h	40.00 (29.00, 50.00)	59.50 (47.75, 67.50)	<0.001
0 h ≥42 [n (%)]	48 (48.50)	20 (99.9)	<0.001
SOFA
0 h	5.00 (2.00, 8.00)	9.00 (7.75, 10.25)	<0.001
0 h ≥7 [n (%)]	37 (37.40)	20 (91)	<0.001
72 h	3.00 (2.00, 5.25)	8.00 (5.00, 12.00)	<0.001
Biomarkers
CRP (mg/L)
24 h	192.00 (142.90, 274.65)	268.00 (186.25, 335.35)	0.070
48 h	228.00 (161.50, 285.50)	241.00 (162.00, 323.00)	0.507
72 h	163.00 (123.00, 247.00)	236.00 (112.85, 313.75)	0.306
Peak	260.60 (188.05, 327.25)	303.00 (232.73, 366.00)	0.116
Lactate (mmol/L)
24 h	1.40 (1.00, 2.00)	1.90 (1.80, 3.90)	<0.001
48 h	1.10 (0.80, 1.50)	1.50 (1.10, 2.60)	0.010
72 h	1.10 (0.80, 1.50)	1.20 (1.00, 1.80)	0.166
Peak	1.70 (1.15, 2.84)	3.10 (2.18, 4.60)	<0.001
Peak ≥1.8 [n (%)]	47 (48)	20 (90)	<0.001
PCT (ng/mL)
24 h	7.24 (2.95, 26.34)	29.68 (6.22, 124.72)	0.012
48 h	5.88 (3.19, 19.53)	25.48 (4.97, 74.37)	0.002
72 h	3.89 (1.84, 16.76)	21.49 (10.88, 56.52)	<0.001
Peak	8.63 (3.25, 27.36)	37.60 (8.47, 145.67)	0.001
Peak ≥100 [n (%)]	4 (4.0)	7 (31)	0.001

SAPS=Simplified Acute Physiology Score; SOFA=Sequential Organ Failure Assessment; CRP=C-reactive protein; PCT=procalcitonin.

The prognostic performance of those variables showing significance in bivariate analysis was assessed by ROC curve analysis, and results for intra-SCCU and overall mortality are shown in Tables 3 and 4, respectively. Significant prognostic performances were observed for SAPS II, SOFA and lactate values, both for intra-SCCU and for overall mortality, and for PCT values (48 h and 72 h levels) for overall mortality.

Table 3.

ROC Curve for Clinical Scores and Biomarkers as Predictors of Intra-SCCU Mortality

	AUC (95% CI)	p
Clinical scores
SAPS II	0.829 (0.686–0.972)	0.001
SOFA
0 h	0.730 (0.599–0.860)	0.021
72 h	0.727 (0.567–0.888)	0.023
Biomarkers
Lactate
24 h	0.831 (0.729–0.932)	0.001
48 h	0.761 (0.623–0.899)	0.009
Peak	0.711 (0.564–0.859)	0.034
PCT
24 h	0.616 (0.393–0.839)	0.246
48 h	0.633 (0.430–0.836)	0.181
72 h	0.668 (0.490–0.847)	0.091
Peak	0.616 (0.416–0.815)	0.246

SCCU=surgical critical care units; ROC=receiver operating characteristics; AUC=area under the curve; CI=confidence interval; SAPS=Simplified Acute Physiology Score; SOFA=Sequential Organ Failure Assessment; PCT=procalcitonin.

Table 4.

ROC Curve for Clinical Scores and Biomarkers as Predictors of Overall Mortality

	AUC (95% CI)	p
Clinical scores
SAPS II	0.829 (0.716–0.942)	<0.001
SOFA
0 h	0.704 (0.572–0.836)	0.017
72 h	0.797 (0.675–0.919)	<0.001
Biomarkers
Lactate
24 h	0.801 (0.695–0.907)	<0.001
48 h	0.687 (0.542–0.831)	0.029
Peak	0.674 (0.537–0.810)	0.042
PCT
24 h	0.632 (0.451–0.813)	0.121
48 h	0.687 (0.517–0.857)	0.028
72 h	0.715 (0.557–0.873)	0.012
Peak	0.650 (0.479–0.821)	0.078

ROC=receiver operating characteristics; AUC=area under the curve; CI=confidence interval; SAPS=Simplified Acute Physiology Score; SOFA=Sequential Organ Failure Assessment; PCT=procalcitonin.

Categorizing variables using as cut-offs values of maximum parsimony in ROC curves, the multivariable analysis for mortality intra-SCCU was significant (R²=0.189, p=0.001), being SAPS II (categorized as high if ≥47) (OR=9.55; 95% CI=1.09–83.85; p=0.042) and lactate value at 24 h (categorized as high if ≥5.87 mmol/L) (OR=6.90; 95% CI=1.28–37.08; p=0.024) associated with mortality. The multivariable analysis for overall mortality was also significant (R²=0.275, p=0.001), with peak PCT levels (categorized as high if ≥100 ng/mL) (OR=11.28; 95% CI=1.80–70.20; p=0.010), peak lactate values (categorized as high if ≥1.8 mmol/L) (OR=8.86; 95% CI=1.51–52.10; p=0.016) and SOFA at admission (categorized as high if ≥7) (OR=8.14; 95% CI=1.69–39.20; p=0.009) being associated with mortality. When high peak PCT and high peak lactate concentrations were introduced as a single dummy variable in the overall mortality model, the model was also significant (R²=0.275, p=0.001), with the dummy variable (OR=99.11; 95% CI=5.21–1885.97; p=0.002) and SOFA at admission (categorized as high if ≥7) (OR=8.16; 95% CI=1.69–39.51; p=0.009) being associated with mortality.

Discussion

Secondary peritonitis is a frequent surgical condition, and clinical scoring systems allow adequate prediction of overall prognosis. On the contrary, biomarkers have been investigated poorly and no specific markers have been identified adequately for assessing overall prognosis of this specific clinical condition.

With respect to scoring systems, whereas the APACHE II score was not helpful in detecting persistent infection in a study on abdominal sepsis [12], but was associated with mortality in others [13,14], the SAPS II score (calculated with data collected during the first 24 h after ICU admission) [15,16] and the SOFA score [17,18] have demonstrated to be good prognostic tools in previous studies. In the present study non-survivors presented greater values of the two clinical scores used, with a significant prognostic performance (AUC >0.700, p<0.05) at admission and 72 h after (in the case of SOFA). The percentage of patients showing values of SAPS II ≥42 or SOFA ≥7 at SCCU admission was greater among non-survivors. SAPS II (OR=9.55, p=0.042) and SOFA at admission (OR=8.14, p=0.009) were also associated with intra-SCCU and overall mortality, respectively, in the multivariable analyses performed.

With respect to biomarkers, CRP, a common marker of systemic inflammation, which increases rapidly in response to trauma and other inflammatory conditions, was not associated with mortality in our study. The lack of differences between CRP values of survivors and non-survivors in our series was not surprising, because all patients had inflammation derived from tissue injury. On the contrary, differences were detected for lactate and PCT, with associations between the lactate value at 24 h and intra-SCCU mortality and between peak lactate and peak PCT values and overall mortality in the multivariable analysis. Peak PCT level was the best biomarker of overall mortality. Both biomarkers have been studied as predictors of mortality in patients with sepsis [19 –22]. Hyperlactatemia has shown to be an independent predictor of mortality in different groups of critically ill patients [23 –25]. In the present study intra-SCCU mortality could be predicted with lactate values at 24 h ≥5.87 mmol/L (OR=6.90, p=0.024) as well as overall mortality with peak lactate values (≥1.8 mmol/L) (OR=8.86, p=0.016).

In the case of PCT, among 17 trials assessing the prognostic value of concentrations with regard to clinical outcome and morbidity in septic patients, 12 trials yielded positive results and five showed negative or equivocal results as reported in a comprehensive search [22]. In the largest study, in patients with intra-abdominal sepsis, median PCT values on days 1, 4, and 10 after surgery were associated with mortality [6]. Other studies also confirmed greater PCT concentrations in non-survivors with peritonitis already in the early postoperative course [4,5]. In our study, PCT levels at 24 h, 48 h, and 72 h were also greater in non-survivors, and peak PCT levels ≥100 ng/mL were associated (OR=11.28, p=0.010) with overall mortality in the multivariable analysis. According to these results, in daily practice critically ill patients with complicated intra-abdominal infection presenting PCT concentrations ≥100 ng/mL can be considered patients who require special attention because they are at greater risk of mortality.

Of interest, in a previous study in patients with sepsis, a concurrent increase in both lactate and PCT concentrations but not clinical severity scores predicted 28-d mortality [26]. In the present study, concomitant high peak lactate and peak PCT concentrations (with the above mentioned cut-offs) considered as a single dummy variable were highly predictive (OR=99.11, p=0.002) of overall mortality, with an odds ratio markedly greater than that for SOFA at admission (OR=8.16, p=0.009). Although ROC AUC values for SAPS II and SOFA scores were higher than for peak lactate and PCT, confidence intervals of clinical scores and biomarkers overlapped, probably because of the non-high number patients in this clinical entity homogeneous study population. However, in the multivariable analysis overall mortality was better predicted by concurrent high peak PCT and peak lactate concentrations than by clinical scores or by each biomarker separately.

The retrospective analysis of the present study represents the major limitation of the study, and prospective clinical trials would be worthy to confirm and validate results in daily practice in this clinical entity. On the contrary, the consideration of a single clinical entity (secondary peritonitis) in the study, the participation of four SCCUs and the large number of patients with this specific entity included, strengthens the results of the study. In this sense it has been postulated that PCT concentrations in one kind of intra-abdominal infection cannot be directly compared with another [27].

Conclusions

The results of the present study suggest that SAPS II and lactate values at 24 h were good predictors of intra-SCCU mortality. However, overall mortality was better predicted by concurrent high PCT and lactate concentrations than by clinical scores or by each biomarker separately. Both PCT and lactate are easily measured and widely available biomarkers, and in our study, when used in combination under routine conditions, provided early prognostic information, more accurate than clinical scores for overall mortality. The results of the study should be observed as a pilot evaluation warranting further prospective studies in this specific clinical entity.

Footnotes

Author Disclosure Statement

Lorenzo Aguilar and Maria-Jose Gimenez are employees of PRISM-AG, which received an investigational grant for data analysis. The remaining authors have no conflicts of interest.

Acknowledgments

We would like to thank Dr. Juan Fernando Diaz for his assistance with the study. Data entry and data analysis of this study was supported by ThermoFisher Scientific (Madrid, Spain).

References

Vincent

, Moreno

, Takala

, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med, 1996; 22:707–710.

Le Gall

, Lemeshow

, Saulnier

. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA, 1993; 270:2957–2963.

Vincent

, Opal

, Marshall

. Ten reasons why we should NOT use severity scores as entry criteria for clinical trials or in our treatment decisions. Crit Care Med, 2010; 38:283–287.

Rau

, Frigerio

, Büchler

, et al. Evaluation of PCT for predicting septic multiorgan failure and overall prognosis in secondary peritonitis: A prospective, international multicenter study. Arch Surg, 2007; 142:134–142.

Schröder

, Staubach

, Zabel

, et al. PCT as a marker of severity in septic shock. Langenbecks Arch Surg, 1999; 384:33–38.

Reith

, Mittelkötter

, Wagner

, et al. Procalcitonin (PCT) in patients with abdominal sepsis. Intensive Care Med, 2000; 26:S165–S169.

Pepys

, Baltz

. Acute phase proteins with special reference to C-reactive protein and related proteins (pentaxins) and serum amyloid A protein. Adv Immunol, 1983; 34:141–212.

Howell

, Donnino

, Clardy

, et al. Occult hypoperfusion and mortality in patients with suspected infection. Intensive Care Med, 2007; 33:1892–1899.

Dandona

, Nix

, Wilson

, et al. PCT increase after endotoxin injection in normal subjects. J Clin Endocrinol Metab, 1994; 79:1605–1608.

10.

Assicot

, Gendrel

, Carsin

, et al. High serum PCT concentrations in patients with sepsis and infection. Lancet, 1993; 341:515–518.

11.

Perren

, Cerutti

, Lepori

, et al. Influence of steroids on procalcitonin and C-reactive protein in patients with COPD and community-acquired pneumonia. Infection, 2008; 36:163–166.

12.

Novotny

, Emmanuel

, Hueser

, et al. Procalcitonin ratio indicates successful surgical treatment of abdominal sepsis. Surgery, 2009; 145:20–26.

13.

Delibegovic

, Markovic

, Hodzic

. APACHE II scoring system is superior in the prediction of the outcome in critically ill patients with perforative peritonitis. Med Arh, 2011; 65:82–85.

14.

Tridente

, Clarke

, Walden

, et al. Patients with faecal peritonitis admitted to European intensive care units: An epidemiological survey of the GenOSept cohort. Intensive Care Med, 2014; 40:202–210.

15.

Chromik

, Meiser

, Hölling

, et al. Identification of patients at risk for development of tertiary peritonitis on a surgical intensive care unit. J Gastrointest Surg, 2009; 13:1358–1367.

16.

Gauzit

, Péan

, Barth

, et al. Epidemiology, management, and prognosis of secondary non-postoperative peritonitis: A French prospective observational multicenter study. Surg Infect, 2009; 10:119–127.

17.

Hernández-Palazón

, Fuentes-García

, Burguillos-López

, et al. Analysis of organ failure and mortality in sepsis due to secondary peritonitis [Article in Spanish]. Med Intensiva, 2013; 37:461–467.

18.

Hynninen

, Wennervirta

, Leppäniemi

, et al. Organ dysfunction and long term outcome in secondary peritonitis. Langenbecks Arch Surg, 2008; 393:81–86.

19.

Riedel

. Procalcitonin and the role of biomarkers in the diagnosis and management of sepsis. Diagn Microbiol Infect Dis, 2012; 73:221–227.

20.

Londoño

, León

, Rodríguez

, et al. Serum lactate in the emergency department as a prognostic factor in patients with sepsis without hypotension [Article in Spanish]. Med Clin (Barc), 2013; 141:246–251.

21.

Suberviola

, Castellanos-Ortega

, Ruiz Ruiz

, et al. Hospital mortality prognostication in sepsis using the new biomarkers suPAR and proADM in a single determination on ICU admission. Intensive Care Med, 2013; 39:1945–1952.

22.

Sridharan

, Chamberlain

. The efficacy of procalcitonin as a biomarker in the management of sepsis: Slaying dragons or tilting at windmills?. Surg Infect, 2013; 14:489–511.

23.

Jansen

, van Bommel

, Woodward

, et al. Association between blood lactate levels, Sequential Organ Failure Assessment subscores, and 28-day mortality during early and late intensive care unit stay: A retrospective observational study. Crit Care Med, 2009; 37:2369–2374.

24.

Mikkelsen

, Miltiades

, Gaieski

, et al. Serum lactate is associated with mortality in severe sepsis independent of organ failure and shock. Crit Care Med, 2009; 37:1670–1677.

25.

del Portal

, Shofer

, Mikkelsen

, et al. Emergency department lactate is associated with mortality in older adults admitted with and without infections. Acad Emerg Med, 2010; 17:260–268.

26.

Phua

, Koay

, Lee

. Lactate, procalcitonin, and amino-terminal pro-B-type natriuretic peptide versus cytokine measurements and clinical severity scores for prognostication in septic shock. Shock, 2008; 29:328–333.

27.

Watkins

, Lemonovich

. Serum procalcitonin in the diagnosis and management of intra-abdominal infections. Expert Rev Anti Infect Ther, 2012; 10:197–205.