Introduction
Materials and methods
Data sources and searches
Study selection
PICOS | Study characteristics |
---|---|
1. Participants | Patients with sepsis and/or septic shock |
2. Intervention | Premorbid exposure to beta blockers |
3. Comparison | No premorbid exposure to beta blockers |
4. Outcomes | Mortality |
5. Study design | Prospective observational or retrospective cohort studies |
Data extraction and quality assessment
Results
Study selection
Characteristics and type of studies
First author | Year of publication | Type of study | Study period (month/year) | Country | Number of centres | Diagnosis | Setting (ED/ICU) | ICU type (medical/surgical) | Outcome | Premorbid beta blocker exposure | Inclusions | Exclusions | Select cohort | No. of patients with premorbid beta blocker use |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Singer et al. [11] | 2017 | Retrospective matched cohort study | 2009–2011 | USA | Medicare Provider and Analysis Review data | Severe sepsis, septic shock | ICU | Mixed | Primary: mortality | Filled prescription within 30 days of admission, including date of admission. Data obtained from Medicare Part D claims data. | Patients admitted with an urgent/emergent hospital admission code, requiring intensive care upon admission, and carrying a primary diagnosis of sepsis or systemic inflammatory response syndrome (SIRS) by ICD9 diagnosis codes, > 65 years, asthma, heart block, 1 year of continuous Part A and Part B coverage, with Part D enrolment. | In hospital claims without ICU admission, Part C enrolment (coverage through healthcare maintenance organizations), beta blocker prescriptions extending into 30 days prior to admission but not through the admission date. | 6839 | 2838 |
Macchia et al. [10] | 2012 | Retrospective matched cohort study | 2003–2008 | Italy | 22 | Sepsis | ICU | Mixed | Primary: mortality | 3 or more filled prescriptions within 4 months of admission. Data obtained from local health authority drug claims database. | Sepsis with codes 038 [septicemia], 020.0 [septicemic], 790.7 [bacteremia], 117.9 [disseminated fungal infection], 112.5 [disseminated candida infection], and 112.81 [disseminated fungal endocarditis]. Admission direct to ICU or transferred to ICU from other hospital departments within 48 h of admission. | < 40 years old, transfer to ICU from other departments 48 h after admission. | 9465 | 1061 |
Hsieh et al. [27] | 2019 | Retrospective matched observational study | 1999–2013 | Taiwan | National Health Insurance Research Database (NHIRD) of Taiwan data | Sepsis, septic shock | – | – | Primary: mortality | Patients were classified as using certain drugs if they took them for more than 1 week within a 3-month period prior to the index date. | First episode of severe sepsis or septic shock. ICD-9-CM coding was used. | Second episode of sepsis. | 33,213 | 1040 |
Fuchs et al. [26] | 2017 | Secondary analysis of prospective observational single-centre trial | 2010–2013 | Germany | 1 | Sepsis, severe sepsis, septic shock | ICU | Surgical | Primary: mortality; survival analysis Secondary: length of stay | Pre-existing oral beta blocker therapy was defined as a treatment started at least 7 days before sepsis onset. | First episode of severe sepsis or septic shock. | < 18 years old, no infection, no chronic beta blocker therapy, no sepsis or septic shock, second episode of sepsis. | 296 | 296 |
Contenti et al. [21] | 2015 | Retrospective cohort study | 2012–2014 | France | 1 | Severe sepsis, septic shock | ED/ICU | – | Primary: initial blood lactate concentration. Secondary: mortality | ‘Premorbid’ not defined. Data obtained from ED observation data or inpatient records | > 18 years old, severe sepsis, septic shock. | – | 260 | 65 |
Sharma et al. [25] | 2016 | Retrospective study | 2013–2014 | ICU | Not specified | Septic shock | ICU | Medical | Peak lactate, cumulative norepinephrine dose and duration, mortality | ‘Premorbid’ not defined. Data obtained from ICU medical records. | Medical ICU, septic shock, vasopressors required. | – | 123 | 48 |
de Roquetaillade et al. [23] | 2018 | Retrospective study | 2008–2016 | France | Not specified | Septic shock | ICU | – | Heart rate, arterial lactate levels, arterial oxygen content, fluid requirements, norepinephrine requirements, duration of mechanical ventilation, mortality | ‘Premorbid’ not defined. | Adult patients diagnosed with septic shock within 48 h. | – | 938 | 230 |
Alsolamy et al. [22] | 2016 | Retrospective cohort study | 1/1/2003–31/12/2013 | Saudi Arabia | 1 | Severe sepsis, septic shock | ICU | – | Primary: mortality | Active prescription 3 months prior to admission. | > 14 years old, severe sepsis and septic shock, previous prescription of beta blockers active for 3 months prior to hospital admission. | – | 4629 | 623 |
Al-Qadi et al. [24] | 2014 | Retrospective study | 2007–2009 | USA | 1 | Severe sepsis, septic shock | ICU | Medical | Primary: mortality | 3 or more months of beta blocker usage prior to admission. Data obtained from electronic records. | Severe sepsis and septic shock, 3 or more months on beta blocker prior to ICU admission. | Patients with comfort care. | 651 | 375 |
Risk of bias assessment
Primary outcome: mortality
First author | Select cohort | No. of patients with no premorbid beta blocker use | No. of patients with premorbid beta blocker use | Mortality census day | Mortality | 90-day mortality | 28-day mortality | ICU mortality | Hospital mortality | Survival analysis | Outcome | Adjustment method | Adjusted variables |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Singer et al. [11] | 6839 | 4001 | 2838 | Hospital mortality | – | – | – | – | aOR = 0.69 (CI 0.62–0.77) | – | Premorbid beta blocker usage is significantly associated with decreased mortality | Multivariate logistic regression | Age, class of beta blocker, congestive heart failure, cancer, surgical procedures |
Macchia et al. [10] | 9465 | 8404 | 1061 | 28-day mortality | – | – | aOR = 0.81 (CI 0.68–0.97), p = 0.025 | – | – | – | Premorbid beta blocker usage is significantly associated with decreased mortality | Multivariate logistic regression | Age, sex, history of hypertension, dyslipidaemia, diabetes mellitus, myocardial infarction, congestive heart failure, atrial fibrillation, chronic obstructive pulmonary disease, depression, and malignancy |
Hsieh et al. [27] | 33,213 | 32,173 | 1040 | Hospital mortality | – | – | – | – | aOR = 0.89 (CI 0.76–1.04), p = 0.1484 | – | Premorbid beta blocker usage is not significantly associated with decreased mortality | Multivariate logistic regression | Age, sex, insurance premium, urbanization level, and comorbidities |
Fuchs et al.a [26] | 296 | 0 | 296 | ICU, hospital, 28 days, 90 days | – | 40.7% vs. 52.7%, p = 0.046a | 28.7% vs. 41.1%, p = 0.04a | 27.5% vs. 38%, p = 0.06a | 35.3% vs. 48.1%, p = 0.03a | HR = 0.67 (CI 0.48, 0.95), p = 0.03a | Continuation of beta-blockade is associated with decreased 28-day, 90-day, and hospital mortality. | Multivariate cox regression | Sex, known nosocomial pathogen, chronic diseases, body temperature (< 36.0 °C), APACHE II score first 24 h, lactate first 24 h (> 3 mmol/L) |
Contenti et al. [21] | 260 | 195 | 65 | 28-day mortality | – | – | – | – | 35% vs 49%, p = 0.08 | – | Premorbid beta blocker usage is not significantly associated with decreased mortality | – | – |
Sharma et al. [25] | 123 | 75 | 48 | Hospital mortality | – | – | – | – | 35.4% vs 32%, p = 0.70 | – | Premorbid beta blocker usage is not significantly associated with decreased mortality | – | – |
de Roquetaillade et al. [23] | 938 | 708 | 230 | ICU mortality | – | – | – | 35.7% vs. 37%, p = 0.75 | – | – | Premorbid beta blocker usage is not significantly associated with decreased mortality | – | – |
Alsolamy et al. [22] | 4629 | 4006 | 623 | ICU mortality | – | – | – | RR = 0.94 (CI 0.82–1.08), p = 0.39 | – | – | Premorbid beta blocker usage is not significantly associated with decreased mortality | – | – |
Al-Qadi et al. [24] | 651 | 276 | 375 | Not specified | 21.3% vs 27.2%, p = 0.09; aOR 0.62, p = 0.023 | – | – | – | – | – | Premorbid beta blocker usage is not significantly associated with decreased mortality | – | Age, gender, and severity of illness using SOFA and APACHE III scores |
Clinical parameters
First author | Heart rate | Initial blood lactate levels | Peak blood lactate levels | Creatinine levels | Arterial pH | Mean arterial pressure | SOFA score | APACHE II score | APACHE III score | Mechanical ventilation | Vasopressor infusion |
---|---|---|---|---|---|---|---|---|---|---|---|
Contenti et al.a [21] | 100 ± 25 vs 109 ± 25 bpm; p = 0.02 | 3.9 ± 2.3 mmol/L vs 5.6 ± 3.6 mmol/L; p = 0.0006 | – | – | – | 72 mmHg ± 22 vs 70 mmHg ± 21; p = 0.48 | 5.0 ± 2.8 vs 5.3 ± 2.8; p = 0.44 | 21.0 ± 6.0 vs 21.7 ± 6.9; p = 0.41 | – | 15% vs 19%; p = 0.58 | 31% vs 32%; p = 0.94 |
Sharma et al.a [25] | – | – | 3.2 vs 3.6 mmol/L; p = 0.54 | – | – | – | – | – | 94 vs 84; p = 0.14 | – | Cumulative dose 11.4 vs 12.6 mg; p = 0.43 Duration of infusion 1563 vs 1730 min; p = 0.37 |
de Roquetaillade et al.a [23] | 81 (IQR 82–111) vs. 107 (IQR 89–122) bpm; p < 0.01 | 1.75 (IQR 0.9–3.4) vs. 1.8(IQR 0.8–4) mmol/L; p = 0.97 | – | 165.5 (IQR 108–245) vs 135.5 (IQR 82–108); p < 0.00 | 7.35 (IQR 7.25–7.42) vs 7.34 (IQR 7.23–7.42); p = 0.354 | – | 9 (IQR 6–12) vs 9 (IQR 6–13); p = 0.242 | – | – | 81.7% vs 84.9%; p = 0.112 Days on ventilation 4 (IQR 2–9) vs 5.5 (IQR 2–11); p = 0.055 | 23.2 mg (IQR 5.1–57.0) vs 22.4 mg (IQR 5.2–60.5); p = 0.95 |
Fuchs et al.b [26] | 111 (IQR 97.0–132.8) vs 118 (IQR 97.0–135.5); p = 0.2 | 2.3 (IQR 1.5–3.8) vs 3.5 (IQR 2.0–6.5); p < 0.01 | – | – | – | – | – | 20.0 (IQR 15.0–24.5) vs 21.0 (IQR 16.2–26.0); p = 0.25 | – | – | Norepinephrine 91% vs 92.2%; p = 0.83 |