Comparison of two optical biometry devices in the measurement of eyes for intraocular lens calculation

Amin Ashrafzadeh, MD and Hossein Zahed, MS

Modesto Eye Center, Modesto, CA, USA

Financial Disclosure:  Amin Ashrafzadeh, MD is a consultant to Carl Zeiss Meditec, Dublin, CA, makers of the IOLMaster.  Hossein Zahed has no disclosures to declare.

Optical biometry systems are used to calculate the IntraOcular Lens (IOL) power for patients undergoing cataract surgery.  IOLMaster and LENSTAR are two such systems that are FDA approved in the United States of America.

Amin Ashrafzadeh, MD and his team at the Modesto Eye Center, Modesto, CA, undertook a study to compare IOLMaster and the LENSTAR instruments.  The following pages and information are the analysis of his data.


All patients underwent complete eye examination by one ophthalmologist (Amin Ashrafzadeh, MD) and were determined to be candidates for cataract surgery in at least one eye.  All patients were phakic bilaterally.

IOLMaster 500 version 7.5, Carl Zeiss Meditec, Inc, Dublin, CA, USA, and the LENSTAR LS 900 version 1.1, Haag-Streit AG, Koeniz, Switzerland, were calibrated at the beginning of each clinic day prior to use.  A sole technician with greater than 10 years of experience in the field of ophthalmology and excellent ease with ophthalmic instruments measured all patients.  All patients were measured using the 2 instruments on the same visit.

The printout from both instruments were collected and tabulated.  Microsoft Excel 2010, Microsoft, Inc., Redmond, WA, USA, and SPSS version 15.0 softwares, SPSS Inc., IBM, Armonk, NY, USA, were used for the analysis of data.  The following items were studied:  Axial Length, Keratometry, Anterior Chamber Depth, Lens Thickness, and the White-To-White.

Additionally, healthy and cooperative, patients without ocular pathology, were enrolled into a separate arm of the study.  This arm of the study was performed at the conclusion of the study to measure the time behind the instrument.  Time was measured from the first interaction of the technician with the instrument to the point when the final measurement was completed on both eyes.


Inability to Capture data with IOLMaster vs. LENSTARA total of 124 eyes of 62 patients were enrolled between February 2012 and June 2012.  The median age of the patients was 70 years with a range of 29 to 89 years, and 24/62 or 38.7% were male and 38/62 or 61.3% were female. 

In the second arm of the study, 20 eyes of 10 patients were enrolled in July 2012 with median age 35 years (range 18-60 years) and 6/10 (60%) were female and 4/10 (40%) were male.

The instruments were not able to capture all the data on all the patients.  In the table to the right, the information on the inability to capture data is presented.  Axial Length (AL), Keratometry (K's), Anterior Chamber Depth (ACD), White-To-White (WTW), and the Lens Thickness (LT) are presented.  It is to be noted, that IOLMaster does not measure lens thickness and therefore, the information is left as blank and since no comparison can be made to the LENSTAR, there is no McNemar test to be performed and that information is left blank as well.

For the remainder of the study, we placed the IOLMaster data as the benchmark, and compared the LENSTAR data relative to benchmark.  Performing the study by placing the LENSTAR as the benchmark and the IOLMaster as the comparison would only achieve a mirror image of the data.


In order to avoid a long page with many graphs, the following pages will discuss the results:

Axial Length (Click Here for Next Page)


Anterior Chamber Depth

Lens Thickness


The efficiency in term of seconds comparing IOLMaster and the LENSTAREfficiency of the Instruments

There 10 patients enrolled for the sole purpose of measuring the time from the interaction with the instruments to the time when the final measurement of both eyes were completed.  The mean time for IOLMaster from the point of data entry to the completion of the measurement was  100.7 seconds or 1:41 (range 85-127 seconds).  The mean time for LENSTAR from the point of data entry to the completion of the measurement was 198.0 seconds or 3:18 (range 176-220 seconds).  The difference of the two means was 97.3 seconds and range for each individual ranged from 74 to 113 seconds.  The P value for this measurement was less than 0.000,000,1.


The LENSTAR was a very difficult instrument to operate and my technician thanked me the day I asked her stop using it.  She found it very difficult, slow and the patients did not appreciate the long 3+ minute examination.  Their eyes dried out and they were uncomfortable.  Just imagine how you would feel sitting with your chin in a machine for 3+ very long minutes!

Final Comments: 

 Axial Length

With only 55.4% (vs. 89.3%) of patients having an axial length measurement of their eyes within 0.3 mm and over 20% (21.3% vs. 3.4%) of patients having greater than 1.0 mm difference in axial length measurement between the right and the left eye with the LENSTAR (vs. IOLMaster), how confident can one be with the axial length measurements by LENSTAR?

With 88% of the eyes measuring within 0.5 D, as a clinician, I wonder if there is any real clinical difference between the keratometry of the two instruments. Future projects should also consider how the inter-instrument correlation of similar make instruments would be. How would keratometric results of two IOLMasters, or two LENSTARs compare?

There is a tremendous amount of hype out there that one instrument is far superior in keratometry and calculations for TORIC IOLs. The current set of data clearly does not support such hype. Aside from "theoretical claims," I await to see real data and science behind the existing claims!

 Anterior Chamber Depth
While still an important element, Anterior Chamber Depth has modest impact on the final IOL power calculation, and there was minimal difference between the two instruments.
 Lens Thickness
It is evident that not having lens thickness available vs. the worst case scenario potential, results in a predicted refractive error of at most 0.08 D for a 29.5 D lens. How clinically relevant is that in your estimation?
The two instruments produced similar means but the range of results by LENSTAR is very concerning. The influence of the White-To-White can be as powerful as the Anterior Chamber Depth. This modest effect has been clearly understated in the clinical circles!
The LENSTAR was a very difficult instrument to operate and my technician thanked me the day I asked her stop using it. She found it very difficult, slow and the patients did not appreciate the long 3+ minute examination. Their eyes dried out and they were uncomfortable. Just imagine how you would feel sitting with your chin in a machine for 3+ very long minutes!

The conclusion of the study is rather self evident.  The Keratometry of the two instruments is nearly identical.  The Anterior Chamber Depth is clinically identical.  The lens thickness is clinically not significant.  The White-To-White results of the LENSTAR are concerning and can have modest effect similar to ACD on the Holladay II results.  The Axial Length is the major issue, where LENSTAR failed, in a major way, to deliver.  Even with best results, I remain very nervous to "hang my hat" on its results.  If the instrument is incapable of delivering the basics, the "fancy" stuff is not really worth considering.  IOLMaster still remains the main, credible, instrument in my practice.