Health & Medical intensive care

Procalcitonin vs CRP for Guiding Antibiotic Therapy in Sepsis

Procalcitonin vs CRP for Guiding Antibiotic Therapy in Sepsis

Materials and Methods

Study Design and Subjects

This was a controlled open randomized clinical trial conducted at two teaching ICUs in Brazil. All adult patients older than or equal to 18 years with suspected severe sepsis or septic shock were assessed for potential inclusion between September 2009 and May 2012. The exclusion criteria, for both groups, were as follows: 1) confirmed microbiological infection by Pseudomonas aeruginosa, Acinetobacter baumannii, Listeria species, Mycobacterium tuberculosis, or fungi; 2) Staphylococcus aureus bacteremia; 3) suspected or confirmed severe infections caused by viruses or parasites; 4) infections that required long-term treatment, regardless of the etiologic agent (e.g., bacterial endocarditis); 5) localized chronic infections (e.g., chronic osteomyelitis); 6) more than 48 hours of antibiotic treatment; 7) immunosuppressed patients (such as those diagnosed with HIV), patients with neutropenia (less than 500 neutrophils/mm), patients post solid-organ transplant, patients under immunosuppressive therapy, and patients who received more than 1 mg/kg of prednisone or equivalent; 8) patients under palliative care; 9) patients who suffered multiple trauma, burns, or major surgery in the previous 5 days; 10) patients diagnosed with pulmonary neoplasias, carcinoid tumors, or medullary tumors of the thyroid; and 11) patients who remained in the ICU for 24 hours or less.

Patients who were eligible for the study were preincluded and monitored for 72 hours before randomization. Patients who did not meet any of the exclusion criteria were randomized to have the duration of antibiotic therapy guided by PCT or by CRP levels.

The local ethics in research board approved the study. All the patients or their guardians signed an informed consent form. This article was written in accordance with the recommendation of the CONSORT statement for Clinical Trials.


The procedures followed to include patients in the study are detailed in Figure 1. Randomization was performed using a table of random computer-generated numbers. Sealed opaque envelopes were used for the randomization.

(Enlarge Image)

Figure 1.

Flowchart for the decision to discontinue antibiotic treatment. PCT = procalcitonin, SOFA = Sequential Organ Failure Assessment, CRP = C-reactive protein.

Serum levels of CRP or PCT were used to encourage discontinuation of antibiotics according to randomization group (Fig. 1). For both groups, the decision to stop antibiotics took into account the clinical response and the behavior of Sequential Organ Failure Assessment (SOFA) score. For patients presenting clinical resolution of infection, a superior limit of 7 days of antibiotic therapy was set, regardless of the CRP and PCT serum levels. Furthermore, patients with positive blood cultures or who had an initial SOFA score > 10 received at least 7 days of antibiotic treatment, even when the remaining discontinuation criteria had been fulfilled (supplemental data, Supplemental Digital Content 1,

The composition of the antibiotic therapy was based on internationally accepted recommendations, and the treating physicians took the final decision regarding when to discontinue treatment.

Measuring PCT and CRP Markers

The reactive test VITROS (Johnson & Johnson Clinical Diagnostics, Inc., Rochester, NY) was used for quantitatively measuring the concentration of serum or plasma CRP. This test has a functional sensitivity of 5 mg/L and linearity between 5 and 90 mg/L. The Vidas BRAHMS PCT (bioMérieux, Lyon, France) was used to measure serum PCT. This test has a functional sensitivity of 0.05 ng/mL and linearity between 0.05 and 200 ng/mL.


The primary outcome was the duration of antibiotic therapy for the first episode of infection. The secondary outcomes were as follows: 1) total number of days on antibiotic therapy; 2) days off antibiotic therapy; 3) death from any cause during the 28 days of follow-up in the hospital; 4) length of stay (LOS) in the ICU and LOS in the hospital; and 5) clinical cure, recurrent infection, and nosocomial infection. Deaths were classified as either related or unrelated to sepsis (see supplemental data, Supplemental Digital Content 2,, for definitions of the secondary outcomes).

Regarding the occurrence of nosocomial infections in the studied patients, courses of antimicrobial therapy administered with an interval of more than 48 hours were considered being directed to distinct infection episodes.

Statistical Analysis

Sample size calculation was based on data from a previous study, in which the mean duration of antibiotic therapy for the index infection was 8.6 ± 5.0 days among patients treated according to a PCT-guided protocol as compared with 10.7 (± 4.0) days in the control group (V. Nobre, unpublished observation, 2008). Thus, we hypothesized that the duration of the antibiotic therapy in patients treated with a PCT-guided protocol would be at least 25% shorter than the duration observed in patients treated according to a protocol based on the serum CRP levels. We found that 58 patients per group—totalizing 116 individuals—would be necessary to demonstrate this difference, with a power of 80% and an alpha error of 5%.

The categorical variables are presented according to their absolute and relative frequency. Regarding the continuous data, the median and the 25–75% interquartile interval (Q1–Q3) were used for the nonnormally distributed variables, whereas the mean and SD were used for the normally distributed variables.

The patients were followed-up for 28 days or until their death or hospital discharge. The occurrence of primary and secondary outcomes was determined by intention-to-treat. Both groups were compared using the chi-square test or Fisher exact test and Student t test or Mann-Whitney U test, as appropriate.

To further investigate the primary outcome, a cumulative antibiotic-discontinuation curve that compared both groups (survival analysis) was created using the log-rank test. Subsequently, Cox proportional hazard model was used to compare the risk of antibiotic discontinuation for the first episode of infection. A severity-adjusted analysis was then performed (Simplified Acute Physiology Score [SAPS] III, Acute Physiology and Chronic Health Evaluation [APACHE] II, and SOFA). The results were displayed using a bivariate analysis with hazard ratios (HRs) and their respective 95% CIs.

A two-tailed test at a significance p value of less than 0.05 was set for all of the analyses. All data were analyzed using the SPSS statistical package, version 15.1 (SPSS, Chicago, IL).

You might also like on "Health & Medical"

Leave a reply