Evaluation of the agreement and reliability of Transpalpebral Tonometers compared with Goldmann Applanation Tonometer – A systematic Review and Meta-Analysis Protocol

Abstract

In 2020, the global prevalence of glaucoma was estimated to be 76 million and it was projected to increase to 111.8 million by 2040. Accurate intraocular pressure (IOP) measurement is imperative in glaucoma management since it is the only modifiable risk factor. Numerous studies have compared the reliability of IOP measured using transpalpebral tonometers and Goldmann applanation tonometry (GAT). This systematic review and meta-analysis aims to update the existing literature with a reliability and agreement comparison of transpalpebral tonometers against the gold standard GAT for IOP measurement among individuals presenting for ophthalmic examinations. The data collection will be performed using a predefined search strategy through electronic databases. Prospective methods-comparison studies published between January 2000 and September 2022 will be included. Studies will be deemed eligible if they report empirical findings on the agreement between transpalpebral tonometry and Goldmann applanation tonometry. The standard deviation and limits of agreement between each study and their pooled estimate along with weights and percentage of error will be reported using a forest plot. Cochrane’s Q test and the I² statistic will be used to assess heterogeneity, and the publication bias will be investigated using a funnel plot, Begg’s and Egger’s tests. The review results will provide additional evidence on the reliability of transpalpebral tonometers that, in turn, could possibly assist practitioners to make informed decision about using it as a screening or diagnostic device for clinical practice, outreach camps, or home-based screening. Institutional Ethics Committee registration number: RET202200390. PROSPERO Registration Number: CRD42022321693.

Glaucoma can be defined in the public health context as a progressive optic neuropathy with characteristic optic nerve head structural changes and functional visual dysfunction often accompanied by typical visual field defects. It is a multifactorial disease and is most often, though not always, associated with increased intraocular pressure (IOP).[1] It is the second leading cause of preventable global blindness among those who are 50 years and older[2] and the third leading cause of blindness globally.[3] The global prevalence of glaucoma in individuals with age between 40 and 80 years is 3.54%. The number of people affected by glaucoma has increased from 64.3 million in 2013 to 76 million in 2020, and it is projected to increase to 111.8 million by 2040.[4] The global distribution of glaucoma is disproportionate, and Asia houses 60% of all patients with glaucoma and 76.7% of those with primary angle-closure glaucoma (PACG).[5]

Elevated IOP is recognized as the most important risk factor for the development and progression of glaucoma.[6] Even though age and family history are considered risk factors, IOP continues to be the only disease-modifying factor that is tractable and predisposed to modulation.[7] A reliable and authentic IOP measurement is imperative since its assessment constitutes the principal screening, diagnosis, and management strategy for patients with glaucoma.[8] Furthermore, there is a 10%–18% increased risk of developing glaucoma for a 1 mmHg increase in IOP, and similarly, an IOP decrease of 1 mmHg curtails the advancement of glaucoma by 10%.[9] This singularly accentuates the importance of accurate IOP measurement for the successful management of glaucoma.

Diverse and distinctive methods of tonometry are in practice for measuring IOP. Since its inception in 1948, Goldmann applanation tonometer (GAT) acquired recognition for its accuracy and reliability in measuring a wide range of IOPs.[10] GAT is still regarded as the “gold standard” that employs a noninvasive measurement technique operating on the Imbert–Fick principle (P = F/A), which states that “the pressure (P) inside an ideal thin-walled sphere is directly related to the force (F) applied to applanate its surface, divided by the flattened area (A).”[11] The precision and repeatability of the device are contingent, assuming a standard central corneal thickness (CCT) of 520 mm (changed to ~ 545 mm for ultrasound optical pachymetry) and a regular corneal composition.[12] GAT measurement is susceptible to discrepancies in CCT[13] and pathological conditions such as corneal curvature, axial length, irregular corneal surface, corneal edema, abnormal corneal curvature, scleral thickness, and rigidity.[14] Besides these deficits, GAT requires anesthesia and is not ideal for resistive patients who are bedridden, in the pediatric setting, as well as for patients who have undergone refractive surgery.[15] Perkins applanation tonometer (PAT) is a portable handheld adaptation of GAT and was proven to provide clinically accepted comparability with that of GAT.[16,17]

Therefore new, portable, noninvasive transpalpebral/eyelid/digital tonometers such as TGDc-01, Diaton, and EASYTON (ET) were introduced to mitigate the measurement error induced by the corneal thickness and other biomechanical properties.[6,18] Numerous research publications have assessed the extent of reliability and applicability of transpalpebral tonometers (TTs) such as TGDc-01,[19] Diaton,[20,21] and EASYTON[22] in clinical practice with variable results. Several publications[15,23] discuss various tonometers in general, and Chakraborty et al.,[6] in their review which compared TT with GAT, concluded that the former overestimated IOP in the lower ranges and underestimated in the higher IOP ranges. In 2012, Cook et al.[18] conducted a meta-analysis which compared the agreement of various tonometers that are in routine use in clinical practice with that of GAT. The study determined a pooled estimate with a mean difference (MD) of − 0.5 mmHg and a 95% limits of agreement (LoA) from − 6.9 to 5.9 mmHg for TTs and 46% of the measurements were estimated to be within the range of 2 mmHg. However, to the best of our knowledge, no meta-analysis has been published recently comparing the reliability of TTs with GAT. Therefore, this systematic review and meta-analysis aims to update the meta-analysis published by Cook et al.[18] by focusing primarily on comparing the reliability and agreement of TTs against the gold standard GAT among individuals presenting for ophthalmic examinations. The current paper describes the methodology employed for conducting the systematic review and meta-analysis.

Methods

Design and registration

A systematic review and meta-analysis study design will be employed to summarize the methods-comparison studies published, which compared TTs with GAT. The development, conduct, design, and reporting of this study protocol are in congruence with the Preferred Reporting Items for Systematic Reviews and Meta-analyses Protocol (PRISMA-P)[24,25] statement and the Guidelines for Reporting Reliability and Agreement Studies (GRRAS).[26] Supplement 1 show the PRISMA-P checklist. This protocol is registered with the International Prospective Register of Systematic Reviews (registration number - CRD42022321693). Even though the current study will use retrospective data, as mandated by our institutional policy, the study with the project code RET202200390 was presented to our Institutional Ethics Committee and was approved on May 30, 2022.

Eligibility criteria

Studies that compare the accuracy of TTs with that of GAT or its equivalent will be considered. A customized search will be conducted to include studies done on human subjects and original research articles published (full text and abstract) in English. When an article is deemed relevant based on the review of its abstract and if the full text is in another language, an attempt will be made to translate the same into English before incorporating it into the review. Studies that used adult participants with normal eyes or with ocular conditions such as glaucoma, ocular hypertension (OHT), and keratoconus will only be included. We will exclude studies that involve children and adolescents (below 18 years). Review articles, case reports, conference abstracts, guidelines, editorials, commentaries, and opinion articles will be excluded. Those studies that compare TTs with other types of tonometers and studies that have used TTs alone without a GAT or PAT comparison will also be excluded.

Search strategy

The data will be gathered employing an electronic search using PubMed, Cochrane Library, and Google Scholar. The search will include a group of terms related to transpalpebral tonometry and IOP measurement. The literature search will include articles published from January 2000 to September 2022. Keywords that will be used in all search engines are (“Tonometer” OR “tonometry” OR “Goldmann Applanation tonometer” OR “contact tonometer” AND “Transpalpebral Tonometer” OR “Digital Eyelid Tonometer” OR “TGDc-01” OR “Diaton” OR “EASYTON”) AND (“Diagnostic accuracy”). Further, in order to ensure the most recent studies are not overlooked, the exact search will be repeated before the final data analysis.

Study selection

The citations ascertained through the search strategy will be exported to Zotero; bibliographic management software and the duplicates will be removed. Two reviewers (NB and JS) will scrutinize the title and abstracts based on predetermined selection criteria, and an inventory of all the selected articles will be created. Any disagreement will be resolved through deliberation and mutual consensus. Full text of all the selected articles will be retrieved and will be read thoroughly to authenticate their appropriateness to be considered for data extraction. The reference lists of all the studies that will be included for analysis will be reviewed to search for any articles that may have been missed during the search. The process of study selection procedure is presented through the PRISMA[27] flow chart [Fig. 1].

Figure 1.

Process of study selection

Data extraction

Data extraction will be performed by two reviewers (JS and VS) using a standard data extraction form that will incorporate essential variables that need to be synthesized from the primary studies. Inconsistency in data extraction between JS and VS will be finalized by the third reviewer BS. To establish whether the agreement between two autonomous reviewers is at random or by chance, Kappa statistics[28] will be used to identify the degree of agreement between the two reviewers.

The following available data will be extracted from the eligible studies using an Excel sheet: the name of the first author and the year of publication, the country in which the study was carried out, study design (retrospective or prospective), the number of eyes included in the study (sample size), mean and standard deviation (SD) of age, mean and SD of IOP measurements for TT and GAT, MD (μ diff), the SD of MD (SD μ diff), LoA (low and high), posture used for obtaining IOP using TT (sitting or supine), and patient preference between the two instruments.

In studies where a single outcome is not reported and where different groups of patients are involved, each group will be considered a separate study. Similarly, in studies that used pre- and post-procedure, each will be considered an independent study. Data will be extracted separately for each of the aforesaid scenarios. Also, if a study has two measurements taken with the same device using two separate enumerators, the measurement taken by the first enumerator will be considered for data extraction.

Statistical analysis

The statistical analysis will be conducted using ReviewManager (RevMan) 5.4 (The Cochrane Collaboration) and StataBE 17 (StataCorp). The risk of bias analysis will be performed using RevMan and the rest of the meta-analysis will be carried out using StataBE 17. The forest plot will be generated to show the individual and pooled SD and 95% LoA, along with the author name, year of publication, study weights, and mean percentage error. Earlier publications on meta-analysis using methods-comparison studies[29,30] have used the framework suggested by Williamson et al.[31] But the current review will attempt to apply a recent framework published by Tipton and Shuster[32] for the meta-analysis of Bland–Altman studies based on the LoA approach.

Assessment of risk of bias

Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2)[33] tool will be used for the assessment of the risk of bias. Quality assessment of bias and applicability has become an indispensable component of the systematic review process which ascertains whether the results of individual studies are valid. The focus is to evaluate the introduction of possible systematic error (i.e., bias) based on the methods adopted, not by the comprehensive assessment of methodological quality. The QUADAS-2 risk of bias evaluation is effected in four major domains, namely, participant selection, index test, reference standard, and flow and timing, which incorporates signaling questions to facilitate risk of bias judgment. Assessment will also be performed to identify applicability concerns that pertain to the first three domains. For each domain and concerns regarding the applicability, studies are rated as “low,” “high,” or “unclear.”[34]

Due to the lack of established guidelines for the quality assessment of method-comparison studies,[31] the assessment question in the QUADAS-2 guidelines will be modified analogously to the approach elaborated by Joosten et al.[30] JS and BS will tailor the quality assessment form to achieve relevance and to make it compatible with the current review comparing the agreement and reliability between TT and GAT. A pilot will be administered to ensure both reviewers achieve consistency in their assessment.

Missing data

The general principles recommended by the Cochrane collaboration will be followed in dealing with the missing data. The corresponding authors of the selected articles will be contacted to request for missing outcome data or, where applicable, for the entire data set to perform further statistical analysis. Pursuant to nonresponse or rejection of data request, data will be imputed based on standardized statistical models. Further, sensitivity analysis will be performed to investigate the robustness and stability of the results, and the potential effect of missing data on review findings will also be discussed.[35]

Testing for heterogeneity

Meta-analysis reporting on heterogeneity has become a standard, and this test explores to ascertain whether the differences between the study findings are due to genuine underlying factors (clinical, methodological or statistical) or due to chance alone.[36] Investigating the existence of coherence among study effects across the included studies in the meta-analysis is imperative since it directly affects the precision with which we can generalize the results of the meta-analysis. The current meta-analysis will use Cochran’s Q test and the I² statistic to assess heterogeneity among individual studies. It will be considered to have statistically significant heterogeneity when the probability value is less than 0.1 (P < 0.1). A low, moderate, or high heterogeneity will be concluded when the value is less than 25%, between 25% and 75%, and more than 75%, respectively.[37]

Assessment of publication bias

The evidence generated by systematic review and meta-analysis, which explains the best available risks and benefits of medical interventions, has the possibility to facilitate better decision-making in the domain of clinical medicine and public health. However, caution must be applied in interpreting the findings which may be influenced by publication bias that can influence the results drastically. Publication bias refers to the tendency or inclination toward the preparation, submission, and publication of study results that are positive or statistically significant findings.[38] The risk of publication bias will be investigated using a funnel plot, which plots bias against the standard error, and further analysis will be performed using Begg’s rank test[39] and Egger’s weighted regression test.[40]

Subgroup analysis

Subgroup analysis and meta-regression will be used to explain possible sources of statistical heterogeneity when there were differences. The following subgroup analyses will be performed based on the availability of relevant data: (i) devices type, (ii) CCT, (iii) time taken between IOP measurement of TT and GAT, (iv) posture used for the IOP measurement with TT, and (v) patient preference between IOP measured using TT and GAT. The MD will be compared with the listed subgroups to draw appropriate conclusions.

Ethics and dissemination

Systematic review deals with secondary data, and therefore, no primary data will be collected directly from human subject participants. Hence, obtaining informed consent is not necessary. The findings of the review will be promulgated in preeminent peer-reviewed journals and will be disseminated in pertinent forums such as conferences and journal clubs.

Review status

The review team has commenced searching for relevant literature in databases based on the inclusion and exclusion criteria mentioned above. We anticipate to complete the review by January 2023.

Conclusion

The available results on the agreement of transpalpebral tonometers compared with GAT show mixed results. No recently published meta-analysis has exclusively compared the agreement and reliability of transpalpebral tonometers with GAT. Therefore, we envisage that our review results might provide additional evidence on the reliability of transpalpebral tonometers. This, in turn, could potentially assist practitioners in making an informed decision about using it as a screening or diagnostic device for clinical practice, outreach camps, or home-based screening. This, we believe, will contribute towards the efforts to enhance early detection and management of glaucoma so that vision loss due to the condition can be prevented to a great extent.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Supplement 1: PRISMA-P Checklist

PRISMA-P Checklist.

Section/Topic	#	Checklist item	Information Reported		Line number(s)
			Yes	No
Administrative Information
Title
Identification	la	Identify the report as a protocol of a systematic review	⊠	□	87
Update	lb	If the protocol is for an update of a previous systematic review, identify as such	⊠	□	80
Registration	2	If registered, provide the name of the registry (e.g., PROSPERO) and registration number in the Abstract	⊠	□	24,25
Authors
Contact	3a	Provide name, institutional affiliation and e-mail address of all protocol authors; provide physical mailing address of corresponding author	⊠	□	Title Page
Contributions	3b	Describe contributions of protocol authors and identify the guarantor of the review	⊠	□	224-229
Amendments	4	If the protocol represents an amendment of a previously completed or published protocol, identify as such and list changes; otherwise, state plan for documenting important protocol amendments	□	⊠
Support
Sources	5a	Indicate sources of financial or other support for the review	□	⊠	NA
Sponsor	5b	Provide name for the review funder and/or sponsor	□	⊠	NA
Role of sponsor/funder	5c	Describe roles of funder(s), sponsor(s), and/or institution(s), if any, in developing the protocol	□	⊠	NA
Introduction
Rationale	6	Describe the rationale for the review in the context of what is already known	⊠	□	67-82
Objectives	7	Provide an explicit statement of the question(s) the review will address with reference to participants, interventions, comparators, and outcomes (PICO)	⊠	□	79-82
Methods
Eligibility criteria	8	Specify the study characteristics (e.g., PICO, study design setting, time frame) and report characteristics (e.g., years considered, language, publication status) to be used as criteria for eligibility for the review	⊠	□	96-107
Information sources	9	Describe all intended information sources (e.g., electronic databases, contact with study authors, trial registers, or other grey literature sources) with planned dates of coverage	⊠	□	109-116
Search strategy	10	Present draft of search strategy to be used for at least one electronic database, including planned limits, such that it could be repeated	⊠	□	112-115
Study Records
Data management	11a	Describe the mechanism(s) that will be used to manage records and data throughout the review	⊠	□	118,119
Selection process	lib	State the process that will be used for selecting studies (e.g., two independent reviewers) through each phase of the review (i.e., screening, eligibility, and inclusion in meta-analysis)	⊠	□	119-126
Data collection process	11c	Describe planned method of extracting data from reports (e.g., piloting forms, done independently, in duplicate), any processes for obtaining and confirming data from investigators	⊠	□	128-133
Data items	12	List and define all variables for which data will be sought (e.g., PICO items, funding sources), any pre-planned data assumptions and simplifications	⊠	□	134-140
Outcomes and prioritization	13	List and define all outcomes for which data will be sought, including prioritization of main and additional outcomes, with rationale	⊠	□	151-152
Risk of bias in individual studies	14	Describe anticipated methods for assessing risk of bias of individual studies, including whether this will be done at the outcome or study level, or both; state how this information will be used in data synthesis	⊠	□	169-174
Data
Synthesis	15a	Describe criteria under which study data will be quantitatively synthesized	⊠	□	134-140
	15b	If data are appropriate for quantitative synthesis, describe planned summary measures, methods of handling data, and methods of combining data from studies, including any planned exploration of consistency (e.g., 12, Kendall’s tau)	⊠	□	176-193
	15c	Describe any proposed additional analyses (e.g., sensitivity or subgroup analyses, meta-regression)	⊠	□	227-229
	15d	If quantitative synthesis is not appropriate, describe the type of summary planned	□	⊠
Meta-bias(es)	16	Specify any planned assessment of meta-bias(es) (e.g., publication bias across studies, selective reporting within studies)	⊠	□	206-211
Confidence in cumulative evidence	17	Describe how the strength of the body of evidence will be assessed (e.g., GRADE)	□	⊠