Antibiotics Reduce Infection But Don't Alter Stroke Outcome

Daniel M. Keller, PhD

October 28, 2014

ISTANBUL, Turkey — Prophylactic ceftriaxone was safe and reduced the rate of infections among stroke patients but had no effect on functional outcomes, a new randomized trial makes clear.

The antibiotic had no effect on functional outcomes among any predefined subgroup but showed some effect in a post hoc analysis of patients who were treated with intravenous (IV) recombinant tissue plasminogen activator (rtPA).

Paul Nederkoorn, MD, PhD, neurologist and clinical epidemiologist at the Academic Medical Center Amsterdam in the Netherlands, postulated that a low number of infections in the study may be a reason for the lack of effect of antibiotics on functional outcomes. Another contributing factor may have been the excellent stroke care that medical centers provide in the Netherlands, where the study was conducted.

Results of the Preventive Antibiotics in Stroke Study (PASS) were presented here at the 9th World Stroke Congress (WSC).

Magnitude of the Problem

Thirty percent of patients develop an infection after stroke, about half of those in the first 48 hours. Pneumonia affects 10% of all stroke patients, and urinary tract infections affect 10%. Infections raise the risk for an overall poor outcome.

Previous studies of preventive antibiotics were relatively small and had other clinical and methodologic problems. A pooled analysis showed that prophylactic antibiotics decreased the infection rate from 36% to 22% without major adverse effects, but their effect on functional outcome remained unclear.

PASS investigators therefore sought to determine whether IV ceftriaxone, 2 g once daily for 4 days, could improve functional outcome in patients with acute stroke. The drug, a broad-spectrum third-generation cephalosporin, is safe, off patent, and inexpensive.

This prospective, randomized, open-label phase 3 trial with a blinded endpoint was carried out in 30 centers. Patients were randomly assigned 1:1 to standard stroke unit care alone or with prophylactic ceftriaxone (n = 1257 completing 3-month follow up in each group). Patients were stratified according to the type of treating center and stroke severity (National Institutes of Health Stroke Scale [NIHSS] score, 1 to 9 vs 10 or greater).

Patients were enrolled between July 2010 and March 2014. Final diagnoses were 83% infarcts, 11% cerebral hemorrhages, 4% transient ischemic attacks, and 2% other diagnoses.

Patients were 18 years or older (median age in the two groups, 73 to 74 years); had an ischemic or hemorrhagic stroke within 24 hours of randomization; and had any measurable neurologic deficit, defined as an NIHSS score of 1 or greater. Among the exclusion criteria were clinical signs of infections requiring antibiotics on admission, antibiotic use within the previous 24 hours, subarachnoid hemorrhage, and imminent death.

Infections were determined by the physician and then scored later by a panel of experts blinded to the treatment allocation, using Centers for Disease Control and Prevention criteria. In case of infection, the treating physician could decide to use additional antibiotics.

The primary endpoint was functional outcome on the modified Rankin Scale (mRS) as assessed by a trained research nurse, blinded to the treatment allocation, using a structured telephone interview. Twelve patients withdrew consent, and 24 were lost to follow-up; these numbers were equivalent between the two study groups. Demographic characteristics were also similar between the groups.

No Difference in Functional Outcome

Treatment with preventive antibiotics had almost no effect on outcomes as measured by mRS. According to an ordinal analysis, there was just about no shift in rankings between the two groups. In both, about 62% of patients had an mRS score of 0 to 2; the rest has a score of 3 to 6.

Safety of the antibiotic intervention was excellent. Two cases of Clostridium difficile diarrhea occurred in the antibiotic group vs none in the control group. As scored by the expert panel, there was one case of ceftriaxone-resistant infection vs none for the controls. Liver enzymes were elevated in 14% of patients in the intervention group and 11% of the control group, but the difference was not statistically significant (P = .08).

Mortality did not differ at discharge (5% in each group) or at 3 months (10% in the intervention group vs 11% for controls).

The length of hospital stay was the same: 6 days (range, 3 to 10 days) for the patients receiving prophylactic ceftriaxone and 6 days (range, 3 to 11 days) for control patients (P = .35).

Regardless of the treatment assignment, infections were strongly associated with unfavorable outcomes.

Table. Diagnosis of Infection by Treatment Group

Infection Ceftriaxone Group, n (%) Control Group, n (%) Odds Ratio P Value
All infections 40 (3) 87 (7) 0.44 <.001
Pneumonia 23 (2) 34 (3) 0.67 .18
Urinary tract 16 (1) 57 (5) 0.25 <.001
Other 5 (0.4) 4 (0.3) 1.25 .75


The only subgroup that showed a significant effect of ceftriaxone administration was patients who had received rtPA (n = 836). Among these, unfavorable outcomes occurred in 33% of ceftriaxone recipients vs 40% of controls (adjusted odds ratio, 0.77; 95% confidence interval, 0.61 - 0.99; P = .04). Given the time window for rtPA administration, patients receiving thrombolysis were treated earlier with an antibiotic than were patients not receiving thrombolysis (8 vs 12 hours, respectively).

However, for all patients treated with ceftriaxone, the time it was begun had no effect on unfavorable outcomes. For 0 to 6 hours, 7 to 12 hours, and 13 to 24 hours, unfavorable outcomes affected 37%, 38%, and 38% of patients, respectively.

Except for thrombolysis, ceftriaxone did not affect unfavorable outcomes for any other subgroup analyzed, including those defined by age (older or younger than 75 years), type of stroke, and stroke severity.

Dr Nederkoorn concluded that prophylactic ceftriaxone is safe and prevents infections but does not affect functional outcomes. Besides the low number of infections, he said the lack of difference in outcome may be attributable to the low number of severe strokes or to a "bystander effect" (ie, infections were merely an indicator but not a cause of poor outcomes).

Observers are now awaiting two additional trials in this area: the STRoke Adverse outcome is Associated WIth NoSocomial Infections (STRAWINSKI) and the Antibiotics to Prevent Infections in Stroke (STROKE-INF).

"Clean and Clear" Study

Session moderator Werner Hacke, MD, PhD, professor and chairman of the Department of Neurology at the University of Heidelberg, Germany, viewed these neutral study results not as a bad thing but as reassuring. He commented to Medscape Medical News that the study results do not differ much from those of earlier, smaller trials.

"I guess what we have seen in the data is that the quality of care in the Netherlands in those centers who participated is so good so that stroke unit care by itself — nursing and early swallowing assessment and so on and so forth — is enough to prevent infections, and you don't have to use antibiotics blindly," he said.

Although ceftriaxone is not expensive, avoiding IV antibiotics is good news because after a few days in a stroke unit, a patient may not need an IV. Thus, getting rid of the IV line "makes things easier," Dr Hacke said.

He noted that the effect was exactly the same regardless of stroke severity, which he also found reassuring.

"It was a clean and clear study, and what Dr [Peter] Sandercock from Edinburgh said [during the question-and-answer period], it is a neutral study, but it gives us very good information about things we don't have to do."

The study did not receive any commercial funding. Dr Nederkoorn, Dr Hacke, and Dr Sandercock have disclosed no relevant financial relationships.

9th World Stroke Congress (WSC). Session MT43: Late Breaking News (no abstract or presentation number). Presented October 25, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.