Infectious Disease · Medical Literature
Procalcitonin to Guide 7 vs 14 Days of Antibiotics in Bloodstream Infections: A Secondary Analysis of the BALANCE Trial.
Abstract
The Bacteremia Antibiotic Length Actually Needed for Clinical Effectiveness (BALANCE) trial showed that 7 days of antibiotics was noninferior to 14 days among patients with bacteremia. However, it is unknown whether patients with high serum procalcitonin (PCT) level at day 7 of therapy would benefit from a prolonged treatment course. To investigate whether an elevated serum PCT level at day 7 of bacteremia was associated with increased mortality among patients treated with 7 vs 14 days of antibiotics. This cohort study was a planned secondary analysis of the BALANCE trial from 2014 to 2023, a multicenter randomized clinical trial where serum was collected on day 7 of bacteremia and patients were followed up for 90 days after study enrollment. Serum samples were collected from patients 7 days after initiation of antibiotics; PCT levels were later quantified by an antibody-based assay and therefore were not available to clinicians. Data were analyzed from January to June 2025. Patients were randomized to receive either 7 or 14 days of antibiotics. Antibiotic selection, dosing, and route were at the discretion of the treating team. The primary outcome was defined as death from any cause within 90 days of the positive index blood culture result. Secondary outcomes included death by any cause while admitted to the ICU, death by any cause while admitted to the hospital, number of days alive and not admitted to the ICU within 28 days of the index positive blood culture, number of days alive and not admitted to hospital within 28 days of the index positive blood culture, and the duration of mechanical ventilation. A total of 125 patients (median age 63, [IQR, 58-79] years; 80 [64.0%] male) were included in this study. Sixty-five participants (52%) had low PCT levels (<250 pg/mL) and 60 (48%) had high PCT levels (≥250 pg/mL) on day 7. The high PCT group was older, had more comorbidities, and had a higher prevalence of community-acquired bacteremia. Ninety-day mortality was higher in the high vs low PCT group: 21.6% (13 of 60) vs 6.2% (4 of 65) (absolute risk difference [ARD] 15.5%; 95% CI, 3.6%-27.5%). Among patients with high PCT level, 90-day mortality was not different among participants with 7 vs 14 days of antbiotics (ARD, -19.9%; 95% CI, -1.7 to 41.7). In this cohort study, elevated PCT level on day 7 was associated with increased 90-day mortality. However, prolonging antibiotics beyond 7 days was not associated with improved mortality in patients with a residually high PCT level. Seven days of antibiotics appears sufficient for most patients with bloodstream infections independent of day 7 serum PCT level.
Clinical relevance
Among hospitalized adults with bacteremia, does an elevated serum procalcitonin (PCT) level at day 7 of antibiotic therapy identify a subgroup of patients who benefit from 14 days of antibiotics compared with 7 days? __SNAPSHOT__ {"studyDesign":"Planned secondary cohort analysis of a multicenter randomized controlled trial (BALANCE trial, NCT03005145); serum PCT measured retrospectively by antibody-based assay — results were not available to treating clinicians at the time of care","setting":"Multicenter international; 74 hospitals across 7 countries; data collected 2014–2023","sampleSize":"n=125 (low PCT [<250 pg/mL]: n=65 [52%]; high PCT [≥250 pg/mL]: n=60 [48%]); parent BALANCE trial enrolled n=3,608","population":"Hospitalized adults with bacteremia enrolled in the BALANCE trial who had stored serum available for PCT quantification on day 7 of antibiotic therapy; community-acquired bacteremia was more prevalent in the high-PCT subgroup, which was also older with more comorbidities","intervention":"7 days of antibiotic therapy (antibiotic selection, dosing, and route at treating clinician discretion)","comparator":"14 days of antibiotic therapy (antibiotic selection, dosing, and route at treating clinician discretion)","primaryOutcome":"All-cause mortality at 90 days after study enrollment","followUp":"90 days","fundingSource":"Not reported in available abstract; some study authors disclosed affiliations with biotech, pharmaceutical, and/or device companies"}
Key findings
Among the 125 patients with available day 7 serum samples, elevated PCT (≥250 pg/mL) was associated with substantially higher 90-day all-cause mortality compared to low PCT (<250 pg/mL): 21.6% vs. 6.2% (absolute risk difference [ARD] 15.5%; 95% CI, 3.6%–27.5%). This difference was statistically significant and represents a clinically meaningful prognostic signal. However, elevated PCT did not identify a treatment-responsive subgroup. Among patients with high PCT (≥250 pg/mL), 90-day mortality was 10.3% in those assigned to 7 days versus 32.3% in those assigned to 14 days of antibiotics (ARD −19.9%; 95% CI, −1.7 to 41.7) — a non-statistically significant difference whose wide confidence interval does not support a benefit of extended therapy and whose direction paradoxically favored shorter treatment. Sensitivity analyses using a higher PCT cutoff of ≥500 pg/mL and after excluding patients on dialysis yielded consistent results: high PCT continued to identify patients at greater mortality risk, but extended antibiotic duration did not improve outcomes. NNT/NNH for extending antibiotic duration in the high-PCT subgroup is not calculable due to lack of statistical significance and the CI crossing zero in both directions. No specific antimicrobial-related adverse event data were reported in the available abstract.
Practice implications
This planned secondary analysis of the BALANCE RCT addresses a clinically important question: does persistently elevated PCT at day 7 identify a subgroup of bacteremic patients who require longer antibiotic courses? The findings indicate that day 7 PCT ≥250 pg/mL is a meaningful prognostic biomarker — patients with elevated PCT had roughly 3.5-fold higher 90-day mortality — but it does not function as a predictive biomarker for antibiotic duration decisions. Extending therapy from 7 to 14 days in high-PCT patients did not confer a survival benefit. These results are directionally consistent with the parent BALANCE trial's noninferiority finding for 7-day treatment. However, several important caveats apply: (1) Only 125 of 3,608 BALANCE participants had stored serum available, raising serious concerns about selection bias and representativeness; the high-PCT subgroup had only 60 patients, making this analysis substantially underpowered to detect realistic interaction effects. (2) PCT assays were performed retrospectively and results were blinded to clinicians, meaning this analysis cannot inform real-time PCT-guided decision-making paradigms. (3) Patients excluded from the parent BALANCE trial (e.g., endocarditis, implanted prosthetic material, Staphylococcus aureus bacteremia per some site protocols, immunocompromised hosts) may represent exactly the population in whom extended therapy remains most clinically relevant — these patients were not adequately represented. (4) The counterintuitive direction of the mortality difference (numerically higher in the 14-day group among high-PCT patients) is almost certainly a reflection of extreme small-sample confounding rather than a true harm of longer treatment. (5) Only the abstract was available for full critical appraisal; important methodological details including adjusted analyses, organism-specific subgroup data, and adverse event reporting require review of the full publication. __LIMITATIONS__ Critically small analytic sample (n=125) drawn from a parent trial of 3,608; the high-PCT subgroup contained only 60 patients, rendering the treatment-by-PCT interaction analysis severely underpowered,Potential selection bias: availability of stored serum was not random and may have systematically differed from patients without stored samples in ways not fully characterizable,PCT assay performed retrospectively and blinded to clinicians — findings cannot be directly extrapolated to prospective, real-time PCT-guided antibiotic stewardship algorithms,The BALANCE parent trial excluded high-risk patients (e.g., endocarditis, deep-seated infections, S. aureus bacteremia at some sites, selected immunocompromised populations), limiting generalizability to these clinically important groups,Confounding by indication is plausible: sicker patients in the high-PCT group may have had their antibiotics extended by treating physicians regardless of randomization, introducing post-randomization bias,Wide and asymmetric confidence interval for the key subgroup comparison (ARD −19.9%; 95% CI, −1.7 to 41.7) precludes definitive conclusions in either direction,Antibiotic selection, dosing, route, and source control were left to clinician discretion, introducing heterogeneity that could confound duration-outcome relationships,Predominantly older, community-acquired, gram-negative bacteremia population (>70% in parent trial); limited representativeness for gram-positive, nosocomial, or polymicrobial infections,Funding sources and potential conflicts of interest not fully disclosed in available abstract,Full-text peer-reviewed publication was not available for complete methodological appraisal at the time of this summary __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ Day 7 serum PCT ≥250 pg/mL identifies bacteremic patients at significantly higher risk of 90-day mortality (21.6% vs. 6.2%) but does not identify a subgroup that benefits from extending antibiotic therapy to 14 days. These findings support the sufficiency of 7-day antibiotic courses for uncomplicated bacteremia regardless of day 7 PCT level, consistent with the parent BALANCE trial. Given the very small analytic sample (n=125), retrospective biomarker measurement, and exclusion of high-risk bacteremia populations, these results should be considered hypothesis-refining rather than practice-changing, and should not be used to justify PCT-guided extension of antibiotic therapy in routine clinical practice without further prospective validation.