Advertisements

S

SAFE

Screenshot 2019-02-22 at 10.01.30.png

Albumin vs. saline for fluid resuscitation

  • Hypothesis-generating subgroup analysis indicated that trauma patients
    may benefit from normal saline whereas septic shock patients may benefit from
    albumin.
  • Included 6997 patients requiring fluid administration to increase intravascular volume.
  • Of the included Patients, 17% trauma, 18% severe
    sepsis
  • Excluded post-op cardiac surgery, liver transplantation, and burns
  • Randomized patients to receive double-blinded 4% albumin or normal
    saline 500 mL boluses.
  • All otheraspects of ICU care were left to the discretion of the treating clinician
  • Primary endpoint (28-day mortality) was not different between albumin and saline (20.9% vs. 21.1%, p=0.87)
  • Duration of mechanical ventilation and ICU length of stay were not different between groups
  • Pre-specified subgroup analysis of 28-day mortality showed a non-significant trend favoring saline in trauma (13.6% vs. 10.0%, p=0-06) and a non-significant trend favoring albumin in severe sepsis (30.7% vs. 35.3%, p=0.09)
  • Post-hoc analysis of trauma subgroup suggested higher mortality with albumin in patients with traumatic brain injury (subgroup N=492,
    mortality 24.5% vs. 15.1%, p=0.009, NNH 11)
  • The overall blinded ratio of albumin:saline volume was 1:1.4 — this study is commonly cited as support that 500 mL albumin is roughly equivalent in resuscitation to 1000 mL normal saline

 


TBL

For all ICU patients requiring fluid resuscitation, albumin was equivalent to
normal saline in 28-day mortality.


See the paper here



 

 

SALT-ED

Screenshot 2019-02-22 at 10.08.33.png

LR vs. Saline in Non-Critically Ill ED Patients

  • Included 13,347 patients in a single center, unblinded, multiple-crossover study design who required at least 500 mL of IV crystalloid and were later admitted to a non-ICU unit of the hospital
  • The emergency department alternated between balanced crystalloids (lactated Ringer’s or Plasma-Lyte A) or normal saline (0.9% NaCl) each calendar month for 16 months
  • There was not a difference in the incidence of KDIGO stage 2 acute
    kidney injury between balanced-crystalloids and saline (8% vs. 8.6%, OR 0.91, 95% CI 0.8 to 1.03, p=0.14)
  • In patients presenting to the emergency department with AKI (KDIGO stage 2 or higher), major adverse kidney events within 30 days; was much lower with balanced-crystalloids than saline (28% vs. 37.6%, NNH 1o, p<0.001)
  • The median crystalloid volume was essentially 1000 mL, indicating that the typical patient likely only received one dose of IV fluid while in the emergency department.
  • The potential renal toxicity of saline is particularly impressive given the relatively small volume of IV fluid administered in this study.
  • Patients receiving balanced-crystalloids instead of saline were less likely to have a low serum bicarbonate (< 20 mmol/L, 24.9% vs. 28%, p<0.01) and less likely to have an elevated serum chloride (> 110mmol/L, 15.2% vs. 19.3%, p<0.01).
  • Balanced-crystalloids (with lactated Ringer’s containing 4 mEq/L of potassium) did not increase serum potassium levels compared to saline (p=0.61).

 


TBL

Among non-critically ill patients in the emergency department requiring IV fluids, balanced-crystalloid IV solutions (Primarily lactated Ringers) does not reduce hospital-free days but may have some impact in reducing major adverse kidney events compared to normal saline.


See the paper here



 

 

SSCOPE

Screenshot 2019-02-22 at 11.13.08.png

Steroids for COPD exacerbations

  • Included 271 patients with COPD exacerbations, age 50+ years, 30+ pack-Year smoking history, and FEV1 < 1.5 L or inability to undergo spirometry due to dyspnea
  • Excluded those with asthma or corticosteroid use in past 30 days
  • Randomized patients within 12 hours of presentation to long-term steroids (8 weeks), short-term steroids (2 weeks), or placebo
  • Corticosteroid regimen: methylprednisolone 125 mg IV Q6hr x 3 days, then prednisone 60 mg (days 4-7), 40 mg (days 8-11), 20 mg (days 12—43), 10 mg (days 44-50), and 5 mg (days 51-57). The “short-term” group stopped on day 15
    without any taper.
  • All patients were hospitalized for at least 3 days and all received 7 days of broad-spectrum antibiotics. For the 6-month study period, all patients were required to use scheduled albuterol, ipratropium, and inhaled triamcinolone
  • Primary endpoint (all-cause mortality, need for mechanical ventilation, COPD readmission, or intensification of pharmacologic therapy) at 30-days was improved with corticosteroids (23% vs. 33%, p=0.04, NNT 10) and 90-days (37% vs. 48%, p=0.04, NNT 9).
  • Importantly, there was no difference between short-term and long-term therapy
  • The composite primary endpoint was heavily driven by “intensification of pharmacologic therapy” (usually initiation of non-blinded corticosteroids)
  • Hospital length of stay was reduced with corticosteroids (8.5 vs. 9.7 days
    p=0.03)
  • There was no difference between placebo and corticosteroids in days hospitalized for COPD after discharge or 6-month all-cause mortality
  • Although not statistically significant, those receiving corticosteroids has more days hospitalized for reasons other than COPD (4.4 vs. 1.2 dayS, p=0.07).
  • Many of these hospitalizations were due to infection/pneumonia
  • Hyperglycemia requiring treatment was more common in patients receiving corticosteroids (15% vs. 4%, p=0.002, NNH 9)

 


TBL

In patients with COPD exacerbations, corticosteroids decreased treatment failure and hospital length of stay but increased hyperglycemia and showed a trend towards more non-COPD hospitalizations. There was no difference between a 2-week and 8-week corticosteroid regimen.


See the paper here



 

 

SEDCOM

Screenshot 2019-02-22 at 11.20.36.png

Dexmedetomidine vs midazolam in mechanically ventilated patients

  • Included 375 mechanically ventilated patients with an anticipated ventilation duration of at least 3 days
  • 60% of patients were CAM-ICU positive at baseline.
  • Randomized to dexmedetomidine 0.8 mcg/kg/hr (titrated 0.2-1.4mcg/kg/hr) or midazolam 0.06 mg/kg/hr (titrated 0.02-0.1 mg/kg/hr) for up to 30 days
  • Blinded loading doses (1 mcg/kg dexmedetomidine or 0.05 mg/kg midazolam) were allowed, but were used less than 10% of the time
  • Open-label midazolam, fentanyl, and haloperidol were allowed (other analgesics/sedatives not permitted)
  • Primary endpoint (time within RASS -2 to +1) was not different between the two groups
  • Dexmedetomidine improved time to extubation (3.7 vs. 56 days, P=0-01), although ICU length of stay was not different
  • Dexmedetomidine reduces the prevalence of ICU delirium (54% vs. 76.7%, p<0.001, NNT 4) and increased delirium-free days (2.5 vs. 1.7 days, p=0.002).
  • This reduction was not associated with a mortality benefit (22.5% vs. 25.4%, p=0.60)
  • Average study drug doses: dexmedetomidine 0.83 mcg/kg/hr and midazolam 0.056 mg/kg/hr (equivalent to 5 mg/hr for 90 kg patient)
  • Open-label midazolam was more common with dexmedetomidine (63% VS_ 49%, p=0.02), although mean midazolam dose per patient was similar
  • Use of open-label fentanyl was not different between groups (opposite result seen in MENDS trial, although MENDS did not allow open-label
    midazolam)
  • Bradycardia (HR < 40) was more common with dexmedetomidine, but incidence of hypotension (systolic BP < 80 mmHg) was not different

 


TBL

Dexmedetomidine was equivalent to midazolam in achieving sedation goals, but there was reduced ICU delirium and duration of mechanical ventilation with dexmedetomidine.


See the paper here



 

 

SEPSIS-pam

Screenshot 2019-02-22 at 11.25.41.png

High versus low MAP goal in septic shock

  • Included 776 patients with septic shock from 29 centers in France
  • Randomized patients to high-target (MAP goal 80-85 mmHg) or low-target (MAP 65-70 mmHg) for up to 5 days
  • High-target patients were reduced to a lower target if an adverse effect occurred that was thought to be due to vasopressor titration (required in 3.6% of high-target patients)
  • Primary endpoint (28-day mortality) was not different between high-target and low-target patients (34% vs 36.6%, p=0.57)
  • Powered to detect an absolute risk reduction of about 10% assuming a primary endpoint of 45%
  • There were no differences in any secondary endpoints, including 90- day mortality, length of stay, or need for mechanical ventilation
  • High-target patients had a higher median dose of norepinephrine on all study days (p<0.001) and a longer duration of vasopressor use (4.7 vs. 3.7 days, p<0.001)
  • Patients with chronic hypertension (about 40% of entire cohort) randomized to high-target therapy were less likely to have a doubling of the serum creatinine (38.9% vs 52%, P=0.02) and less likely to require renal replacement therapy (31.7% vs 42.2%, p=0.046)
  • High-target therapy was associated with a higher risk of atrial fibrillation (6.7% vs. 2.8%, p=0.02)
  • High-target therapy was associated with a higher risk of atrial fibrillation (6.7% vs. 2.8%, p=0.02)

 


TBL

In patients with septic shock, a higher MAP goal did not reduce mortality but did increase the risk of atrial fibrillation. In a pre-specified subgroup of patients
with a history of hypertension, a higher MAP goal was associated with a reduction in the need for renal replacement therapy.


See the paper here



 

 

SHOCK
Screenshot 2019-02-22 at 11.32.14.png

Early revascularization versus medical stabilization in cardiogenic shock

  • Included 302 patients with cardiogenic shock following a myocardial infarction (STEMI, new left bundle branch block, posterior infarction with anterior ST depression)
  • Randomized patients to early revascularization (angioplasty or bypass within 6 hours of randomization) or medical stabilization
  • Primary endpoint (30-day mortality) was no different between revascularization and medical-therapy (46.7% vs. 56.0%, p=0.11).
  • The trial was adequately powered to detect a 20% absolute difference between
    groups.
  • Six-month mortality was lower in the revascularization group (50.3% vs. 63.1%, p=0.027, NNT 8)
  • As part of pre-specified subgroup analysis, revascularization was more beneficial in younger patients (< 75 Yrs old) and those with prior MI
  • Acute renal failure (defined as a creatinine > 3 mg/dL) was more common with medical-therapy (13% vs. 24%, p=0.03)
  • Revascularization patients received angioplasty (54.6%) or bypass graft (375%) a mean of 12 hours after the onset of MI symptoms
  • Protocol compliance was very high.
  • Within the revascularization group, 97% underwent coronary angiography and 87% received revascularization.
  • Similarly, only 2.7% of medical stabilization patients were revascularized within 54 hours of randomization.
  • Thrombolytic therapy was common (49.3% revascularization vs. 63.3% medical), which limits the external validity of the study given that lytic therapy is now uncommonly done in North America

 


TBL

Compared to medical stabilization, emergent revascularization in cardiogenic shock did not improve 30- day mortality, but did confer a survival benefit at six months. The benefit of early revascularization was most
pronounced in younger patients with
previous MI.


See the paper here



 

 

SIC

Screenshot 2019-02-22 at 11.38.00.png

Selenium in intensive care

  • Included 249 patients from 11 ICUs in Germany with APACHE III scores 2 70 and at least two SIRS criteria.
  • Following diagnosis of SIRS, patients needed to be enrolled within 24 hours and begin treatment within one hour of inclusion.
  • Of the 249 randomized patients, 11 withdrew for a variety of reasons, which formed the 249 intent-to-treat (IT—l) patients. Of the I’l’l’ group, 49 patients (21%) were excluded due to protocol violations (typically drug administration failure), which formed the 189 per-protocol (PP) patients.
  • Randomized patients to sodium- selenite 1000 mcg bolus, then 1000 mcg/day (2 mL/hr) continuous infusion for 14 days (total of 6.9 mg elemental selenium) or placebo (48 mL normal saline per day)
  • Importantly, the study was designed using a one-tailed significance of a = 0.025 (NOT the traditional two-tailed test with a = 0.05)
  • Primary endpoint (28-day mortality) in the I’l’l’ group did not show a difference in mortality between selenium and placebo (39.7% vs. 50.0%, p=0.109)
  • There was a trend toward lower 28- day mortality in the PP analysis with selenium (42.4% vs. 56.7%, p=0.049); however this p value was above the significance level of 0.025)
  • In predefined subgroup analyses, more critically ill patients tended to demonstrate a mortality benefit than less sick patients (eg, elevated APACHE III score, more than three organ failures, sepsis with declining platelet count, and septic shock)
  • The study’s adverse effect analysis was questionable and not robust.
  • Adverse effects were common, but no specific adverse effect was attributed to selenium in this study.
  • The benefit of selenium is hypothesized to involve selenium- dependent glutathione-peroxidases (GPX), an enzyme responsible for redox reactions. GPx-3 activity is negatively correlated with severity of sepsis.
  • In this trial, serum selenium and GPx-3 levels were higher in patients receiving selenium treatment.
  • The study was significantly underpowered, used a one-tailed analysis, and relied on a per-protocol analysis for demonstrating a trend toward efficacy.
  • To date, this is the most robust selenium trial for ICU Patients, but due to methodologic flaws, the efficacy of selenium is still controversial.

 


TBL

Selenium supplementation was not associated with a decrease in mortality among a heterogeneous ICU population with SIRS criteria; however, a favorable trend was noted.


See the paper here



 

 

SLEAP

Screenshot 2019-02-22 at 11.56.01.png

Light sedation with and without daily interruption

  • Included 423 patients (most non-surgical) expected to require 48+ hours of mechanical ventilation
  • Randomized patients to protocolized sedation plus daily interruption (“interruption” group) or protocolized sedation alone (“control”)
  • Analgesic and sedation medications could include an opioid (morphine, hydromorphone, fentanyl) or a benzodiazepine (midazolam, lorazepam).
  • Propofol and dexmedetomidine were NOT permitted.
  • Daily interruption of sedation was performed similarly to the original Kress 2000 article — infusions were held until patients became agitated or followed commands, and then restarted at 50% of the previous rate
  • Primary endpoint (time to extubation) was no different between the two groups (median 7 days in both groups, p=0.52)
  • The trial met 90% power to detect a 2 day extubation difference with a baseline time to extubation of 7 days
  • There were no differences in other clinical outcomes between the two study groups (eg, ICU length of stay, mortality, unintended removal of devices, delirium, reintubation)
  • Mean sedation scores were similar between the two groups; “Interruption” patients received higher average midazolam equivalents (4-25 mg/hr vs. 3.4 mg/hr, p=0.04)
    and fentanyl equnvalent (74 mcg/hr vs. 45 mcg/hr, p<0.001)
  • The use of a sedation interruption occurred in 72.2% of eligible “interruption” patient study days (a compliance rate higher than previous
    studies)
  • There is debate why this trial is in direct conflict with the Kress 2000 and ABC trials.
  • This trial used a nursing-driven, protocolized sedation strategy that targeted a lighter sedation strategy than the other trials.
  • Although the trial did not allow propofol (commonly used and
    recommended sedative), the benefit of sedation interruption should be
    most prominent with benzodiazepines

 


TBL

In mechanically ventilated patients with a light sedation strategy, daily interruption of sedation did not improve patient outcomes. In fact, interruption was associated with higher opioid and benzodiazepine requirements.


See the paper here



 

 

SMART

Screenshot 2019-02-22 at 13.19.01.png

LR vs. Saline in Critically Ill ICU Patients

  • Included 15,802 patients in a single center, unblinded, multiple-crossover study design who were admitted to one of five intensive care unit of the hospital
  • Each ICU alternated between balanced crystalloids (lactated Ringer’s or Plasma-Lyte A) or normal saline (0.9% NaCl) each calendar month for 12 months
  • The trial had a pragmatic design –the treating clinician determined the
    close of crystalloid and all aspects of the patient’s care. An electronic alert was provided for balanced-crystalloids to consider avoiding their use in patients with hyperkalemia or brain injury (recommending saline instead)
  • Primary endpoint (major adverse kidney events within 30 days) was lower with balanced-crystalloids than saline (14.3% vs. 15.4%, OR 0.9, 95% CI 0.82 to 0.99, NNH 91, p=0.04)
  • “Major adverse kidney events” was defined as death, new renal replacement therapy, or a final serum creatinine 2 200% of baseline value
  • Balanced-crystalloids was numerically (but not quite statistically) superior with regards to 30-day mortality (10.3% vs. 11.1%, p=0.06), the need for renal replacement therapy (2.5% vs. 2.9%, p=0.08), and the rate of KDIGO stage 2 AKI or worse (10.7% vs. 11.5%, p=0.09)
  • In a subgroup of patients with sepsis (about 15% of the study), 30-day mortality was lower with balanced- crystalloids than saline (25.2% vs- 29.4%, OR 0-8, 95% CI 0.67 to 0.97, P=0-02) — this should be considered a hypothesis-generating finding as there were many subgroup analyses conducted
  • Between balanced-crystalloids and saline, there was no difference in the rate of the final serum creatinine 200% of baseline (6.4% vs. 6.6%, p=0-6) or ICU-free clays (median 25.3 vs. 25.3, p=0.94)
  • The median cumulative crystalloid volume was essentially 1000 mL in both groups, indicating that the typical patient only received one dose (1 L bag) of IV fluids while in the ICU.
  • The potential renal toxicity of saline is particularly impressive given the relatively small volume of IV fluid administered in this study. The study did not specifically analyze a dose-response effect demonstrating that larger doses of crystalloid magnified the effect on the primary endpoint.
  • Patients receiving balanced- crystalloids instead of saline were less likely to have a low serum bicarbonate (< 20 mmol/L, 35.2% vs. 42.1%, p<0.001) and less likely to have an elevated serum chloride (> 110 mmol/L, 24.5% vs. 35.6%, p<0.001).
  • Balanced-crystalloids (with lactated Ringer’s containing 4 mEq/L of potassium) did not increase serum potassium levels based on similar peak serum potassium levels and similar instances of serum potassium exceeding 5 mmol/L.
  • The results of this trial are in agreement with SALT-ED, suggesting that even smaller closes (about 1 L) of saline may cause harm compared to balanced-crystalloids, particularly as it relates to the rate of renal toxicity.

 


TBL

Among critically ill patients, balanced- crystalloid IV solutions (lactated Ringers or Plasma-Lyte A) reduces major adverse kidney events (the composite of death, renal replacement therapy, or doubling of baseline serum creatinine) compared to saline.


See the paper here



 

 

SOAP-II

Screenshot 2019-02-22 at 13.25.54.png

Dopamine vs norepinephrine for shock

  • Included 1679 patients requiring vasopressor support for shock despite a small fluid challenge (60% septic, 20% cardiogenic, 15% hypovolemic)
  • Randomized to dopamine (titrated by 2 mcg/kg/min to max 20 mcg/kg/min) or norepinephrine (titrated by 0.02 mcg/kg/min to max 0.19 mcg/kg/min — 15 mcg/min for 80 kg patient)
  • Open-label norepinephrine was allowed once blinded vasopressor reached maximum dose
  • Primary endpoint (28-day mortality) was not different between the two groups (52.5% vs. 48.5%, p=0.10).
  • ICU, hospital, 6-month, and 12-month mortality rates were also not different between groups
  • Cardioqenic shock, a pre Specified subgroup, showed higher 28-day mortality with dopamine (p=0.03)
  • Dopamine group required more open- label norepinephrine (26% vs. 20%, p<0.001), had higher urine output in first 24 hours (but similar overall fluid balance), and had a higher heart rate
  • Arrhythmias, mostly atrial fibrillation, were more common with dopamine (24.1% vs. 12.4%, p<0.001, NNH 9)

 


TBL

Among patients with all types of shock, mortality rates were not different between norepinephrine and dopamine, although norepinephrine was more effective as a vasopressor and was less associated with arrhythmias. Norepinephrine may have a mortality benefit over dopamine in a subset of patients with cardiogenic shock.


See the paper here



 

 

SPLIT

Screenshot 2019-02-22 at 13.30.50.png

Plasma-Lyte 148 vs. 0.9% NaCl for IV fluid

  • Included 2278 patients in four New Zealand ICUs who required IV crystalloid fluid.
  • Patients requiring renal replacement therapy were excluded.
  • Patients in the study had very few comorbidities and were primarily admitted to the ICU after elective surgical Pprocedutres-
  • Very few patients with a high likelihood of mortality were included.
  • Randomized each ICU to “PL-148” (Plasma-Lyte, a buffered crystalloid solution) or “NS” (0.9% normal saline) in alternating blocks of 7 week periods over a total of 28 weeks
  • Primary endpoint (acute kidney injury within 90 days) was not different between PL—148 and NS (9.6% vs. 9.2%, p=0.77)
  • If a difference in the primary endpoint does exist, the study ruled out an absolute percentage harm of >2.1% or a benefit of >2.9% with PL-148 versus NS
  • All other endpoints related to acute kidney injury, length of stay, and mortality were not different between the two groups
  • The median amount of fluid received was 2000 mL for both PL—148 and NS, which was mostly given within the first 24 hours of ICU admission
  • Because patients in this trial did not receive large quantities of crystalloid fluid (median 2000 mL), it is unlikely that the hyperchloremic “toxicity” of normal saline would be evident.
  • Further, the study did not include a significant number of patients who are most at-risk for AKI (eg, septic shock, severe trauma, etc).

 


TBL

Among a very heterogeneous ICU patient population, the incidence of acute kidney injury was not different between relatively small crystalloid volumes of Plasma-Lyte 148 compared to normal saline.


See the paper here



 

 

STOP-IT

Screenshot 2019-02-22 at 13.35.57.png

Short-course antimicrobial therapy for intraabdominal infection

  • Included 518 patients with complicated intraabdominal infection who had adequate source control
  • Randomized in open-label fashion to control (antibiotics lasting two days after the resolution of fever, leukocytosis, and ileus up to 10 days) or a short antibiotic course regardless of the clinical picture (4 days)
  • Primary endpoint (composite of surgical—site infection, recurrent intraabdominal infection, or death within 30 days) was similar between control and short course (22.3% vs. 21.8%, p=0.92)
  • The individual endpoints of surgical- site infection, recurrent intraabdominal infection, death, Clostridium diffici/e infection, extraabdominal infection with resistant pathogens, and hospitalization were not different between the two groups;
  • As expected, control group patients had a longer duration of antimicrobial therapy (median 8 days vs. 4 days, p<0.001)
  • Compliance with the study protocol was less common among the control group (72.7% vs. 81.8% compliance, p=0-02)- In both groups, protocol violations were mostly due to prolonged antibiotic durations.
  • Although recurrent intraabdominal infections were equally likely to occur between the groups, patients in the control group on average had recurrent infections occur later in the Postoperative course (15.1 vs. 10.8 days, p<0.001), indicating that antibiotics delayed but did not prevent recurrent infections
  • 30-day mortality was quite low within the study (1%), which may indicate that sicker patients (who may benefit from longer antimicrobial durations) were not enrolled
  • The study was stopped early for futility and funding after enrolling about half of the planned patients and was significantly underpowered.
  • Statistically, the study demonstrated that if a difference does exist, the difference in the primary endpoint is either an absolute 7% in favor of a short course or an absolute 8% in favor of a longer course of antimicrobial therapy.
  • While the authors discuss that the difference of less than 10% is not clinically meaningful, many clinicians would find such a difference important (eg, a primary endpoint rate of 20% vs. 150/0).

 


TBL

Among patients with complicated intraabdominal infection and adequate source control, a shortened antibiotic duration of 4 days was not different in surgical-site or recurrent intraabdominal infections versus a longer course (about 8 days as guided by clinical symptoms).


See the paper here



 

 

SORT 1999

Screenshot 2019-02-22 at 13.45.52.png

Albumin for spontaneousbacterial peritonitis

  • Included 126 patients with cirrhosis and primary SBP (spontaneous bacterial peritonitis), defined as an ascitic PMN count > 250
  • Randomized to cefotaxime/placebo or cefotaxime/albumin. Albumin 20% was given as 1.5 gm/kg at enrollment and 1 gm/kg on day 3
  • Primary endpoint (renal impairment) was lower in patients receiving albumin (33% vs. 10%, p=0.002, NNT 4)
  • In-hospital mortality (29% vs. 10%, p=0.01, NNT 5) and 90-day mortality (41% vs. 22%, p=0.03, NNT 1:5) were both significantly reduced in the albumin group
  • Resolution of infection, duration of antibiotics, and hospital stay were not different between groups
  • Criticisms of study are that it was not blinded and there is no data regarding the fluid management of the placebo group. It is not known whether lower” doses or fluid expansion with less expensive alternatives (normal saline) would produce the same beneficial effects

 


TBL

Two doses of concentrated albumin in patients with spontaneous bacterial peritonitis reduced renal impairment and mortality compared to placebo.


See the paper here



 

Advertisements

Powered by WordPress.com.

Up ↑

%d bloggers like this: