Methods

Study design: parallel, multicentre, single‐masked (participant), randomised controlled trial

Unit of randomisation: participant

Participants

Country: conducted at 12 sites in the United States and 8 international sites

Total number of participants enrolled: 152 participants (152 eyes)
Total number of participants randomised: 152

Number of men and women: women 54.7% (Hydrus group), 58.4% (iStent group)

Age range: men: 66.9, SD 10 (Hydrus group), 66.5, SD 9.5 (iStent group)

Inclusion criteria:

• Phakic or pseudophakic

• A diagnosis of POAG treated with hypotensive medications

• Medicated IOP ≤ 31 mmHg

• Diurnal IOP ≥ 23 mmHg and ≤ 39 mmHg

Exclusion criteria:

• Congenital or developmental glaucoma

• Previous trabeculectomy or other glaucoma procedure, argon laser trabeculoplasty

• Ab interno or ab externo device implanted in or through Schlemm's canal

• Use of oral hypotensive medication for glaucoma for treatment of fellow eye

Interventions

Intervention: Hydrus microstent (N = 75)
Comparator: iStent (n.2) trabecular micro‐bypass stent (N = 77)

Outcomes

Primary outcome
Proportion of participants unmedicated at 12 months following surgery (taken from ClinicalTrials.Gov)

Secondary outcomes

• Mean change in unmedicated IOP from baseline to 12 months (wash‐out was not possible for some participants)

• Mean medication use at 12 and 24 months post procedure

• Surgical success, defined as freedom from secondary surgery, IOP 18 mmHg or less, and discontinuation of all ocular hypotensive medications (taken from report)

• Visual field testing using the 24‐2 SITA standard strategy using a Humphrey Visual Field Analyzer (Carl Zeiss Meditech, Jena, Germany) was collected at baseline, 3 and 12 months, but not listed as an outcome measure in ClinicalTrials.Gov and in the published article.

Safety outcomes

• Intraoperative complications

• Observed rate of ocular adverse events

Length of follow up: 12 months

Notes

Date conducted: Participants were randomised from March 2013 to May 2015.

Funding source: Study sponsored by Ivantis, Inc., Irvine, California

Declaration of interest: several study authors had received honoraria, grants, consulting fees from Ivantis Inc., as well as other companies.

Trial ID: NCT02023242

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: "The allocation was determined by a computer generated sequence stratified by site and prepared in advance by the study statistician in order to provide balanced study groups"

Allocation concealment (selection bias)

Low risk

Quote: "Randomization was performed in the operating room by opening a sequentially numbered envelope"

Comment: not enough details on how the process was managed.

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Participants were masked

Blinding of outcome assessment (detection bias)
All outcomes

High risk

Quote: "the investigator at each study site was not masked to treatment randomization during follow‐up examinations."

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Two participants were lost in each group

Selective reporting (reporting bias)

Low risk

There appears to be no major difference between the published report and details on the protocol in ClinicalTrials.gov; however, visual field was obtained but not reported.

Methods

Study design: parallel, multicentre, single‐masked (participant), randomised controlled trial

Unit of randomisation: participant

Participants

Country: conducted at 26 sites in the United States and 12 international sites

Total number of participants enrolled: 1143
Total number of participants randomised: 556 participants (556 eyes)

Number (%) of men and women: women 55.8 (intervention group), 56.1 (comparator group)

Age range: mean 71.1, SD 7.9 (intervention group), 71.2, SD 7.6 (comparator group)

Inclusion criteria:

• An operable age‐related cataract

• A diagnosis of POAG treated with 1 to 4 hypotensive medications

• Medicated IOP ≤ 31 mmHg

• Diurnal IOP ≥ 22 mmHg and ≤ 34 mmHg

Exclusion criteria:

• Congenital or developmental glaucoma

• Previous argon laser trabeculoplasty

• Ab interno or ab externo device implanted in or through Schlemm's canal

• Use of oral hypotensive medication for glaucoma for treatment of fellow eye

Interventions

Intervention: Hydrus microstent + CS with phacoemulsification (N = 369)
Comparator: CS with phacoemulsification only (N = 187)

Outcomes

Primary outcome

• Proportion of eyes at 24 months with unmedicated mean MDIOP reduction ≥ 20% compared with baseline (taken from study report)

Secondary outcomes

• Mean change in unmedicated MDIOP from baseline to 24 months

• Changes in mean medication count per participant between baseline and 24 months follow‐up.

• Proportion of eye medication free at each visit.

Safety outcomes

• Intraoperative complications

• Observed rate of ocular adverse events

Length of follow up: 24 months

Adverse events reported: Yes

Notes

Date conducted: Participants "were assessed for study eligibility between February 2012 and April 2015". Study first completion date was June 2017

Sources of funding: Study sponsored by Ivantis, Inc., Irvine, California

Declaration of interest: Several study authors had received honoraria, grants, consulting fees from Ivantis Inc., as well as other companies.

Trial ID: NCT01539239

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

"Upon confirmation, eyes were randomized to either Hydrus Microstent implantation (HMS group) or no microstent implantation (NMS group) using an online computer algorithm in a 2:1 allocation ratio."

Allocation concealment (selection bias)

Low risk

"Upon confirmation, eyes were randomized to either Hydrus Microstent implantation (HMS group) or no microstent implantation (NMS group) using an online computer algorithm in a 2:1 allocation ratio."

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

"Study subjects remained masked to treatment assignment throughout the course of the study."

Blinding of outcome assessment (detection bias)
All outcomes

Unclear risk

"The tonometry protocol utilized a 2‐person method: an observer and a reader who was masked to study treatment."

"Despite multiple measures to minimize bias, it was not possible to mask the surgeon to treatment group during postoperative examinations."

Incomplete outcome data (attrition bias)
All outcomes

Unclear risk

"After accounting for the randomization ratio and 10% annual attrition, the study size was calculated to be 558 subjects."

556 participants were randomized (so trial investigators did try to avoid attrition bias). 3% were lost to follow‐up and 2% died or could not return owing to non‐study‐related critical illness. No details given on losses in each study arm or methods used to account for missing data.

Selective reporting (reporting bias)

Unclear risk

There appears to be no major differences between the published report and details on the protocol in ClinicalTrials.gov; however, worsening of visual field was not listed as an outcome measure, but was obtained since visual field loss by 2.5 dB or more was reported as an adverse event (4.3% for Hydrus and 5.3% for iStent).

Methods

Study design: Parallel, multicentre, randomised, single‐masked (participant), controlled clinical trial

Unit of randomisation: participant

Only 1 eye per participant was eligible for treatment, although both eyes could be screened for inclusion.

"Before surgery, participants were washed out of all hypotensive medications in the study eye for a variable period, depending on the class of medication in use at the time of screening. The washout protocol is described in the Ocular Hypertension Treatment Study. At the completion of the washout, a preoperative baseline diurnal IOP (DIOP) value was obtained by averaging 3 Goldmann tonometry measurements obtained 4 hours apart between 8AM and 4PM. The tonometry protocol used a 2‐person system (an observer and a reader), and 2 readings were obtained at each time point during the day. If the difference in the 2 measurements was more than 2 mmHg, a third measurement was obtained. The average of 2 measurements or the median value of 3 was used for the time point, and the average of the IOP measurements at all 3 time points was the mean DIOP. The DIOP value was required to be between 21 and 36 mmHg for study inclusion."

Participants

Country: study conducted at 7 European sites: Germany, Spain, the Netherlands, Italy

Total number of participants randomised: N = 100

Number of men and women: 40% men (intervention Group), 58% (comparator group).
Average age and age range: 21 to 80 years old; mean 72.8, SD 6.6 (intervention group); 71.5, SD 6.9 (comparator group)

Inclusion criteria: People with concurrent cataract and open‐angle glaucoma ("IOP of 24 mmHg or less with no more than 4 hypotensive medications, Shaffer grade III or IV chamber angle in all quadrants and Humphrey visual field changes characteristic of glaucoma or glaucomatous optic nerve damage confirmed by ophthalmoscopy and nerve fiber layer imaging").

Exclusion criteria: "Clinical exclusion criteria included angle‐closure glaucoma, secondary glaucomas except pseudoexfoliation or pigment dispersion syndromes, exudative age‐related macular degeneration (AMD), proliferative diabetic retinopathy, or significant risk of glaucomatous vision loss because of washout of IOP‐lowering medications. Anatomic exclusion criteria were narrow angle or other angle abnormality visible on gonioscopy, central corneal thickness of less than 480 mm or more than 620 mm, or clinically significant corneal dystrophy. participants with prior corneal surgery, argon laser trabeculoplasty, cycloablation, or any incisional glaucoma procedure, such as trabeculectomy, tube shunts, deep sclerectomy, or canaloplasty, also were excluded."

Interventions

Intervention: Hydrus microstent + CS with phacoemulsification (Hydrus + CS) N = 50
Comparator: CS with phacoemulsification (CS) only N = 50

Outcomes

Primary outcome

• Proportion of participants with a 20% or more reduction in mean washed‐out diurnal IOP at 12 and 24 months.

Secondary outcomes

• Mean washed‐out diurnal IOP

• Proportion of participants taking hypotensive medications

• Proporton of participants using medication throughout the follow‐up period

• Number of glaucoma medications at follow‐up (24 months)

Safety outcomes

• Intraoperative complications

• Observed rate of ocular adverse events

• Change in visual acuity

• Secondary glaucoma surgery: 1 (Hydrus + CS), 2 (CS)

Length of follow up: 24 months

Intervals at which outcomes assessed: Follow‐up examinations were conducted per protocol at 1 day, 1 week, and 1, 3, 6, 12, 18, and 24 months.

Loss to follow‐up
"Before the 12‐month visit, 2 participants from the Hydrus + CS group and 1 participant from the CS group exited the study for non‐health‐related reasons, for a 12‐month subject
accountability rate of 97 (97%) of 100. Between 12 and 24 months, 4 additional participants exited from the study: 1 participant died of cardiac disease, 1 participant developed lung cancer, 1 declined further participation after secondary glaucoma surgery, and 1 participant was lost to follow‐up, all in the CS group, for a 24‐month accountability rate of 93 (93%) of 100."

Adverse events reported: Yes

Notes

Date conducted: Participants randomised to the study from July 2011 to April 2012

Sources of funding: Study sponsored by Ivantis, Inc., Irvine, California

Declaration of interest: The trial investigators have declared their financial disclosures in the trial report including financial support from Ivantis Inc.; Transcend; Glaukos, Innfocus and Alcon.

Trial registration: NCT01818115

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Quote: ". . . were assigned randomly in a 1:1 ratio according to a computer‐generated listing just before surgery"

Allocation concealment (selection bias)

Low risk

Quote: ". . . were assigned randomly in a 1:1 ratio according to a computer‐generated listing just before surgery"

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Quote: "Subjects remained masked to treatment assignment for the course of the study."

Comment: Single‐masked study where the personnel were not masked.

Blinding of outcome assessment (detection bias)
All outcomes

High risk

"Masking the surgeon to the assigned treatment was not possible, and because the microstent is visible on the slit lamp with gonioscopic examination, masking the treatment group from the IOP assessor during follow‐up visits also was not possible."

Incomplete outcome data (attrition bias)
All outcomes

Low risk

All participants lost to follow‐up were accounted for.

Selective reporting (reporting bias)

Low risk

Outcomes match those reported on ClinicalTrials.gov; however, Humphrey visual field was collected but not listed as an outcome measure and not reported in the manuscript.