1. Initiële behandeling: literatuur samenvatting

Study reference

Study characteristics

Patient characteristics

Intervention (I)

Comparison / control (C)

Follow-up

Outcome measures and effect size

Comments

Navarro, 2016

Type of study: randomised, double-blind, phase 3 trial 

Setting: 13 emergency medical service centres and 26 hospital departments.

Country: France

Source of funding: UCB Pharma.

Inclusion criteria: patients aged at least 18 years, with GCSE duration of more than 5 min or generalised convulsions with no recovery of consciousness between seizures

Exclusion criteria: women with known or clinically detected pregnancy; patients with known allergies to clonazepam, levetiracetam, or pyrrolidone; and patients whose status epilepticus was linked to a pathological condition, such as trauma, who needed immediate surgery, patients with post-anoxic or subtle status epilepticus (defi ned by minor and erratic myoclonic movements in patients with severely impaired consciousness), those whose parent, guardian, or other reliable person refused permission and those not covered by the French medical insurance system, who should not be included in clinical trials according to the French Public Health Code. Additionally, patients with definite non-epileptic psychogenic pseudo-status, patients who previously received another drug for the ongoing status epilepticus, or those who had been included in the study previously for an episode of status epilepticus.

N total= 125

Intervention group:

N= 68

Mean age ± SD: 55(18)

Sex: 72% M (n=49)

Control group:

N= 68

Mean age ± SD: 53(18)

Sex: 66% M (n=45)

Groups comparable at baseline? yes

Clonazepam (1mg) with levetiracetam (2.5g)

Patients were given 1 mg clonazepam by intravenous perfusion for 1 min, concurrent with 2·5 g levetiracetam or placebo intravenous perfusion for 5 min. If the status epilepticus continued after 5 min, a second dose of 1 mg clonazepam was injected.

Clonazepam (1mg) with placebo

Endpoint of follow-up: seizure cessation within 15 min of the first injections of clonazepam and levetiracetam or placebo.

For how many participants were no complete outcome data available? N=11

Reasons for incomplete outcome data described? Yes

Significant differences between groups? No

Seizure cessation within 15 min of the onset of treatment:

I: 50/68 (74%)

C: 57/68 (84%)

RR: 0·88 (0·74–1·05); p=0,14

Time between the first injection and cessation of convulsions, min:

I: 3 (0–50)

C: 5 (0–41)

P=0.97

Funded by UCB Pharma

Gujjar, 2017

Type of study: Prospective open-label randomized study 

Setting: Patients presenting with status epilepticus (SE) or cluster attacks of seizures (CS) to the emergency department of the university-teaching hospital in Oman were recruited.

Country: Oman

Source of funding: This study was funded by Dean’s Fund, College of Medicine and Health Sciences, Sultan Qaboos University, Muscat, Oman. 

None of the authors have any conflicts of interests to declare concerning this study. 

Inclusion criteria: Consecutive adult (> 15 years age) patients presenting with SE or CS to our university-teaching hospital were recruited. SE is defined for the purpose of this study as a prolonged (> 5 min) or recurrent generalized tonic-clonic seizure(s) with no regaining of consciousness between attacks, or partial seizures persisting for more than 10 mins. Cluster attacks of seizures (CS) were defined as recurrent episodes of seizures two or more over previous 24 h (partial or generalized) with return of conscious- ness between episodes, with the last episode occurring within 2 h of evaluation.

Exclusion criteria: Patients with known allergy to drugs used, those requiring immediate neurosurgery, having hemodynamic compromise (systolic BP <100 mm Hg), serious arrhythmias (eg. heart rate <50/min, second- or third-degree heart block, runs of ventricular ectopics), pregnancy, cardiac failure or pulmonary edema, in pre-terminal states or having pseudo-seizures were excluded. Patients with subtle SE, defined as altered mental status without overt convulsions but with ongoing epileptiform activity noted on EEG.

N= 52 patients with SE (M/F: 34/18, mean age: 37.8±18 years) and

63 patients with CS (M/F: 44/19, mean age 39.3±20 years)

LEV group SE:

N= 22 (42%)

Mean age ± SD: 38±19

Sex: M/F: 13/9

DPH group SE:

N= 30 (58%)

Mean age ± SD: 37±19

Sex: M/F: 21/9

LEV group CS:

N= 38 (60.3%)

Mean age ± SD: 38.5±19

Sex: M/F: 26/12

DPH group CS:

N= 25 (39.7%)

Mean age ± SD: 40.6±21

Sex: M/F: 18/7

Groups comparable at baseline? A higher proportion of patients in DPH group required initial IV benzodiazepine as compared to the LEV group (76% vs 40%; p = 0.004). 

Intravenous levetiracetam administered after an IV benzodiazepine, 20 mg/kg over 30 min

Intravenous phenytoine administered after an IV benzodiazepine, 30 mg/kg over 30 min

Endpoint of follow-up: NR

For how many participants were no complete outcome data available? NR

Reasons for incomplete outcome data described? NA

Significant differences between groups? NA

Rates of seizure control over 24h:

SE: LEV was effective in 18/22 (82%) while DPH was effective among 21/30 (70%) (p= 0.33). 30 patients received LEV as first or second agent for SE; of these, 23/31 (76.6%) were controlled. Similarly, 32 patients received DPH as first or second AED and 22/32 (68.8%) were controlled. When sequentially used, the two anticonvulsants were effective in controlling SE among 48/52 (92.3%) patients.

CS: EV was effective as the first AED among 31/38 patients (81.6%) while DPH was effective among 20/25 (80%). When used as either first or second AED, LEV and DPH were effective among 36/43 (83.7%) and 22/29 (75.9%) of the cases respectively. With the use of one or both these AEDs, CS were controlled in 61/63 (96.8%) patients; only two patients were resistant to both AEDs.

Adverse effects:

SE: Rates of adverse events were low. Transient hypotension (responding to IV saline) was observed in two patients receiving IV DPH; one patient on LEV had transient thrombocytopenia which normalized over three days, while another had transient agitation and anxiety.

CA: Adverse event rates were low, with two patients receiving DPH experiencing local pain, while transient agitation was noted in one patient in LEV group. Good outcome rates were higher in the LEV group compared to DPH (79% vs 70%); however, these differences were only marginally significant (p = 0.049). One patient with a progressive encephalopathy of suspected autoimmune etiology and sepsis who received DPH for cluster seizures died after prolonged ICU care. On univariate analysis, age, plasma albumin, total proteins and anticonvulsant used (favoring LEV) were significantly associated with outcome status at discharge (p > 0.05). Multiple logistic regression analysis revealed plasma protein level (p = 0.03, Exp B: 0.9; 95% CI: 0.83–0.99) and anticonvulsant used (p = 0.004; ExpB:0.018; 95% CI: 0.001–0.27) to be factors independently predicting outcome. 

Conclusion: IV Levetiracetam controls status epilepticus or cluster seizures with an efficacy comparable to that of phenytoin. Use of these two agents consecutively may control >90% of all such conditions without resort to anaesthetic agents. Further studies should explore its efficacy in larger cohorts of epileptic emergencies.

Limitations: Randomization failed in 1/3 of the patients due to urgency of required interventions.

Small number of patients with varying underlying causes 

Randomization failed in 1/3 of the patients due to urgency of required interventions.

Small number of patients with varying underlying causes

Su, 2016

Type of study: Prospective randomized controlled trial

Setting: emergency room or neurocritical care unit of the Xuanwu Hospital, Capital Medical University

Country: China

Source of funding: This study was supported by the National Key Department of Neurology funded by the Chinese Health and Family Planning Committee, and the National Key Department of Critical Care Medicine funded by the Chinese Health and Family Planning Committee.

Su Yingying, Liu Gang, TianFei, Ren Guoping, Jiang Mengdi, Brian Chun, Zhang Yunzhou, Zhang Yan, Ye Hong, GaoDaiquan, and Chen Weibi declare they have no conflicts of interest. 

Inclusion criteria: All adult patients (aged 18 years or older) with GCSE who were admitted to the emergency room or neurocritical care unit of the Xuanwu Hospital, Capital Medical University, initially received first-line treatment with diazepam (intravenous injection of 0.2 mg/kg, administered twice at a 10-min interval) according to hospital protocol but not responding to this first-line treatment.

Exclusion criteria: (1) unstable vital signs, such as systolic blood pressure \90 mmHg, pulse \60 beats/min, or arterial blood oxygen saturation \90%; (2) liver dysfunction (alanine transaminase or total bilirubin more than twice the upper limit of normal); (3) neurologic emergency requiring immediate surgical intervention; (4) pregnant or breast- feeding; (5) hypersensitivity to study drugs

N=73, mean age:41.72 ± 17.14 years (range 18–80)

Etiology of GCSE:

viral encephalitis (n = 30), pre-existing epilepsy (n = 23), other undetermined causes (n = 14), and cerebrovascular disease (n = 6).

Intervention group:

N=37

Mean age ± SD: 37.14 ± 14.98

Sex: M/F: 19/18

Duration of GCSE:

<4h: 8 (21.62%)

>4h: 29 (78.38)

Control group:

N=36

Mean age ± SD: 45.26 ± 18.14

Sex: M/F: 20/16

Duration of GCSE:

<4h: 10 (27.78%)

>4h 26 (72.22%)

Groups comparable at baseline? Yes

Intravenous pheno-barbital administered after IV diazepam.

The loading dose of 20 mg/kg (an additional 5–10 mg/kg may be administered) began at a rate of 50 mg/min, followed by an intravenous dose of 100 mg every 6 h.

Intravenous valproate administered after IV diazepam.

The loading dose of 30 mg/kg (an additional 15 mg/kg may be administered) began at a rate of 3 mg/kg/min, followed by a continuous infusion at a rate of 1–2 mg/kg/h. The infusions were maintained for 24–48 h and then gradually tapered until they were eventually replaced with oral AEDs (24–72 h). 

Endpoint of follow-up: 3 months

For how many participants were no complete outcome data available? None

Reasons for incomplete outcome data described? NA

Significant differences between groups? NA

Effective seizure control (confirmed by a certified neurologist and EEG):

I: 81.1% (30/ 37)

C: 44.4% (16/36);

OR 5.357, 95% CI 1.869–15.356; p= 0.001)

Relapse rate of SE <24h (I/C):

6.7%, 2/30 vs. 31.3%, 5/16) [Chi-square 4.888, OR 0.157,95% CI 0.026–0.934; p= 0.040

Relapse rate NCSE <24h (I/C): 20.0% (6/ 30) vs. 31.3% (5/16) (Chi-square 0.726, OR 0.550, 95% CI 0.138–2.197; p= 0.477)

Severe adverse effects (I/C): 13,5% versus 0%, p=0,004

The mortality (phenobarbital vs. valproate) rates were 8.1% (3/37) vs. 16.6% (6/36) at discharge, and 16.2% (6/37) vs. 30.5% (11/ 36) at 3 months. Trends for the risk of developing post-SE symptomatic epilepsy were 26.3% (5/19) and 42.8% (6/14) for the phenobarbital and valproate groups, respectively.

Like other studies, a high level of termination rate was seen with phenobarbital, therefore a lower intravenous rate was chosen in this study design.

Small number of patients.

Single centre Chinese population

Amiri-Nikpour, 2017

Type of study: RCT, using block randomization

Setting: 110 consecutive patients suffering from benzodiazepine refractory SE who were referred to the emergency ward from March 2014 to March 2015 and were randomly divided into two groups

Country: Iran

Source of funding: This study was funded by the Research Council of Urmia University of Medical Sciences, Urmia, Iran.

Inclusion criteria: patients aged at least 18 years, with generalized convulsive seizure lasting greater than 5 min or to or more discrete seizures during which the patient had not returned to baseline consciousness

Exclusion criteria:

Pregnant women, patients younger than 18 years, patients with history of liver diseases, patients

requiring emergency neurological invasive interventions, and finally patients suffering from hypotension, pancreatitis, congestive heart failure, cardiac arrhythmias, postanoxic SE, nonepileptic seizures, or sensitivity to phenytoin or sodium valproate were all excluded from the study.

N= 110

Intervention group:

N= 55

Mean age ± SD: 42.16 (15.94)

Sex: 43.6% M (n=24)

Control group:

N= 55

Mean age ± SD: 43.69(17.60)

Sex: 43.60% M (n=27)

Groups comparable at baseline? Yes

Intravenous sodium valproate (Depakine; Sanofi- Aventis), 30 mg/kg as the loading dose and then 8 hr as maintenance regimen

Intravenous phenytoin (Hydantoic; Caspian Tamin), 20 mg/kg as loading dose and then 1.5 mg/kg every 8 hr as maintenance therapy.

Before being labelled as RSE all patients had been treated by IV diazepam in doses of 0.2 mg/kg at 2 mg/min up to a maximum 20 mg.

Endpoint of follow-up: 7 days regarding drug response and adverse effects

For how many participants were no complete outcome data available? N=0

Reasons for incomplete outcome data described? n/a

Significant differences between groups? No

Seizure control within 7 days of administration

I: 43/55 (78.18%)

C: 39/55 (70.90%)

p=0.428

RR: not reported. Self-calculated: RR=1.10 (0.89 – 1.37); p=0.38

Seven-day mortality

I: 7/55 =12.73%

C: 7/55 =12.73%

RR=1

P=0.612

This study was supported by the Research Council of Urmia University of Medical Sciences, Urmia, Iran.

Author’s conclusion

Sodium valproate is preferred to IV PHT for treatment and control of SE due to its higher tolerability and lower hemodynamic instability

Not blinded

Study reference

Design

Inclusion Criteria

Quality

Results

Comments

Prasad, 2014

Systematic review and meta-analysis (Cochrane review)

Inclusion criteria: (quasi-)randomised controlled trials comparing the effectivity of anticonvulsant drug against placebo or another anticonvulsant drug to people with premonitory, early, established or refractory status epilepticus (convulsive as well as non-convulsive)

Exclusion criteria: none

Search: adequate

Selection: adequate

Quality assessment of individual studies: yes, there were a few studies with high risk of selection or performance bias. For selection, performance and detection bias there were studies with an unclear risk.

Adults with generalized convulsive status epilepticus:

Chen, 2011: valproaat IV vs diazepam IV à

Non-cessation: 15/30 vs. 16/36, RR 1.13 (0.67-1.88)

Hypotension: 0/30 vs. 2/36, RR 0.24 (0.001-4.79)

Relapse of seizure: 3/15 vs. 5/20, RR 0.8 (0.23 - 2.83)

Misra, 2011: levetiracetam IV vs lorazepam IV à

non-cessation: 9/28 vs. 10/41, RR 0.97 (95%CI 0.44-2.13)

seizure recurrence at 1-24h: 6/29 vs. 10/31, RR 0.64 (95%CI 0.27-1.54)

Hypotension: 2/23 vs. 8/21, RR 0.23 (95%CI 0.05-0.96)

Respiratory failure: 5/23 vs. 10/21, RR 0.46 (95%CI 0.19-1.12)

Death: 10/23 vs. 9/21, RR 1.01 (95%CI 0.51-2.00)

Ventilatory need: 4/23 vs. 10/21, RR 0.37 (0.13-0.99)

Shaner, 1988: diazepam in combination with phenytoin IV vs phenobarbitone IV à

Non-cessation: 8/18 vs. 2/18, RR 4.0 (95%CI 0.98-16.3)

Adverse effects: 9/18 vs. 9/18, RR1.0 (95%CI 0.77-1.3)

Ventilatory support: 6/18 vs. 6/18, RR 1.0 (95%CI 0.4-2.52)

Death: 0/18 vs. 0/18, RR 0.0 (95%CI -0.1 – 0.1)

Silbergleit, 2012: midazolam IM vs lorazepam IV à

Seizure-cessation: 329/448 vs. 282/445, RR 1.16 (95%CI 1.06-1.27)

Endotracheal intubation: 63/448 vs. 64/445, RR0.98 (95%CI 0.71-1.35)

Hospitalisation: 258/448 vs. 292/445, RR0.88 (0.79-0.97)

ICU admission: 128/448 vs. 161/445, RR 0.79 (0.65-0.96)

Seizure recurrence: 51/448 vs. 47/445, RR 1.08 (0.74-1.57)

Adverse effects: 16/448 vs. 18/445, RR 0.88 (0.46-1.71)

Mean length of ICU stay: 5.7 vs. 4.1 days, mean difference 1.6 (95%CI -0.23 – 2.43)

Mean length of hospital stay: 6.7 vs. 5.5 days, mean difference 1.2 (95%CI -0.24 – 2.64)

Adults with refractory SE:

Agarwal, 2007, looking at patients refractory to diazepam IV, comparing valproaat IV vs phenytoin IV. à

Non-cessation: 6/50 vs. 8/50, RR 0.75 (95%CI 0.28-2.00)

Adverse effects: 4/50 vs. 8/50, RR 0.5 (95%CI 0.16-1.55)

Deaths: 4/50 vs. 4/50, RR 1.00 (95%CI 0.26-3.78)

Rosetti, 2011: propofol IV vs. barbiturate IV (pentobarbital in USA OR Thiopental in Switzerland) à

SE controlled with first course of drug: 6/14 vs. 2/9, RR 1.93 (0.49-7.54)

SE treated subsequently: 4/8 vs. 5/7, RR 0.7 (0.3-1.62)

Thrombotic/embolic complications: 0/14 vs 0/9, RR not estimable

Mortality: 6/14 vs. 3/9, RR 1.29 (0.43-3.88)

Functional outcome at 3 wks: 5/14 vs. 3/9, RR 1.07 (0.34-3.42)

Infections requiring antibiotics: 7/14 vs. 6/9, RR 0.75 (0.37-1.51)

Children with status epilepticus – within hospital:

Ahmad, 2006: Intranasal lorazepam vs IM paradehyde à

Seizure-cessation: 60/80 vs. 48/80, RR 1.22 (95%CI 0.99-1.52)

Death: 15/80 vs. 13/80, RR1.15 (95%CI 0.59-2.27)

Silbergleit, 2012: midazolam IM vs lorazepam IV à

Seizure-cessation: 329/448 vs. 282/445, RR 1.16 (95%CI 1.06-1.27)

Endotracheal intubation: 63/448 vs. 64/445, RR0.98 (95%CI 0.71-1.35)

Hospitalisation: 258/448 vs. 292/445, RR0.88 (0.79-0.97)

ICU admission: 128/448 vs. 161/445, RR 0.79 (0.65-0.96)

Seizure recurrence: 51/448 vs. 47/445, RR 1.08 (0.74-1.57)

Adverse effects: 16/448 vs. 18/445, RR 0.88 (0.46-1.71)

Mean length of ICU stay: 5.7 vs. 4.1 days, mean difference 1.6 (95%CI -0.23 – 2.43)

Mean length of hospital stay: 6.7 vs. 5.5 days, mean difference 1.2 (95%CI -0.24 – 2.64)

Sreenath, 2010: diazepam in combination with IV phenytoin vs IV lorazepam à

Non-cessation: 0/88 vs. 0/90, RR 0.0 (95%CI -0.02 – 0.02)

Adverse effects: 5/88 vs. 4/90, RR 0.78 (95%CI 0.22 – 2.82)

Death: 0/88 vs. 0/90, RR not estimable

Ventilatory support: 0/88 vs. 0/90 RR not estimable

Children with status epilepticus – at home: N.R.

Children with refractory status epilepticus:

Mehta, 2007: valproate IV vs diazepam IV à

Non-cessation: 4/20 vs. 3/20, RR 1.33 (0.34-5.21)

Hypotension: 0/20 vs. 10/20, RR 0.05 (0.00-0.76)

Respiratory depression: 0/20 vs. 10/20, RR 0.05 (0.00-0.76)

Singhi, 2002: midazolam IV vs. diazepam IV à

Non-cessation: 3/21 vs. 2/19, RR 1.36 (95%CI 0.25 – 7.27)

Ventilatory support: 11/21 vs. 9/19, RR 1.11 (95%CI 0.59 – 2.07)

Adverse effects: 8/21 vs/ 9/19, RR 0.8 (95%CI 0.39-1.66)

Death: 8/21 vs/ 2/19, RR 3.62 (95%CI 0.87-14.97)

Brigo, 2016a

Meta-analysis

Randomized controlled trials comparing IV LEV or IV VPA against IV PHT and RCTs comparing IV LEV versus IV VPA used as second-line treatment for convulsive SE (generalized or focal) were included in the meta-analysis.

All RCTs, blinded or not blinded, were included and excluded were uncontrolled and nonrandomized trials. Trials were not excluded on the basis of dose, duration of treatment, or length of follow-up.

Patients from any age group diagnosed with convulsive SE persisting despite first-line AEDs (benzodiazepines) were included.

Search: adequate

Selection: adequate

Quality assessment of individual studies: most studies had an unclear risk for allocation bias (selection bias) due to insufficient information. Also some studies have high risk in random sequence generation. 

Four studies were included (with a total of 321 episodes of SE), three comparing IV VPA with IV PHT and two comparing IV LEV with IV PHT or with IV VPA and IV PHT. One study compared the efficacy of three different AEDs (VPA, PHT, LEV).

IV VPA versus IV PHT:

3 studies: 227 episodes of SE. No significant statistical heterogeneity among trials was detected. Therewas no statistically significant difference in clinical seizure cessation after drug administration between the IV VPA and the IV PHT groups (OR: 1.07; 95% CI: 0.57 to 2.03).

IV LEV versus IV PHT:

2 RCTs: 144 episodes of SE. No significant statistical heterogeneity in these RCTs. There was no statistically significant difference in clinical seizure cessation after drug administration between the IV LEV and the IV PHT groups (OR: 1.18; 95% CI:0.50 to 2.79).

IV LEV versus IV VPA:

1 meta-analysis: After drug administration, SE was controlled in 78% (39 out of 50) patients receiving LEV and in 68% (34 out of 50) patients treated with VPA. The difference was not statistically significant (P N 0.05).

Common reference-based indirect comparisons by combining meta-analyses of AEDs (IV LEV versus IV VPA, common comparator PHT): No statistically significant difference was found between IV LEV andIV VPA in clinical seizure cessation after drug administration (OR: 1.16; 95% CI: 0.45 to 2.97).

Brigo, 2016b

Systematic review with meta-analysis of RCTs

Inclusion criteria: Patients from any age group diagnosed with SE, clinically defined as convulsive seizures lasting more than 5 min. Both generalized and focal convulsive subtypes of SE were included. All RCTs, blinded or not blinded. No language restrictions.

Exclusion criteria: Focal SE cases without enough information on the presence of convulsive activity as well as nonconvulsive SE were excluded. Uncontrolled and nonrandomized trials were excluded. 

Search: adequate

Selection: adequate

Quality assessment of individual studies: except for 1 RCT, risk of bias was reported to be low. 

Five RCTs were included (656 patients, 320 randomly allocated to IV lorazepam; 336 to IV diazepam).

Seizure cessation after drug administration:

5 RCTs, 587 episodes. No statistically significant heterogeneity was found; no statistically significant differences were found: RR=1.09; 95%CI 1.00-1.20.

Similar results found in subgroup-analysis with only adults or children.

Continuation of SE requiring a different drug:

5 RCTs, 587 episodes. No statistically significant heterogeneity was found; no statistically significant differences were found: RR=0.76; 95%CI 0.57 - 1.02

Similar results found in subgroup-analysis with only adults or children.

Seizure cessation after a single dose of medication:

3 RCTs, 392 episodes. No statistically significant heterogeneity was found; no statistically significant differences were found: RR=0.96; 95%CI 0.85 - 1.08.

Similar results found in subgroup-analysis with only adults or children.

Need for ventilator support:

4 RCTs (585 episodes). No statistically significant heterogeneity was found; no statistically significant differences were found: RR= 0.93; 95%CI 0.61 - 1.43.

Similar results found in subgroup-analysis with only adults or children.

Wu, 2016

Meta-analysis

Inclusion criteria: Prospective RCTs, patients with generalized CSE that underwent treatment with lorazepam of diazepam to control their status epilepticus.

Exclusion criteria: Not mentioned

Search: adequate

Selection: adequate

Quality assessment of individual studies: adequate: all included trials scored with satisfactory quality with Jadad Scale and CONSORT checklist

Six studies (n=970 patients of which 574 children, 396 adults) with level II evidence were included. All studies were assessed to be of high quality.

Seizure control in adults: (n=3 trials): diazepam versus lorazepam: no significant difference (odds ratio [OR], 0.73, 95% CI 0.35–1.55; I2 = 0.0%, p = 0.157).

Seizure control in children: (n=3 trials): no significant difference found between diazepam and lorazepam: OR, 0.95, 95% CI 0.73–1.22; I2 = 0.0%, p = 0.677.

Adverse effects in adults: (n=3 trials): no significant difference between diazepam and lorazepam: OR, 1.13, 95% CI 0.73–1.74; I2 = 0.0%, p = 0.580.

Adverse effects in children: (n=3 trials): no significant difference between diazepam and lorazepam: OR, 1.35, 95% CI 0.92–1.99; I2 = 0.0%, p = 0.129.