Effect of Cataract Surgery on Optical Coherence Tomography Measurements and Repeatability in Patients With Non-Insulin–Dependent Diabetes Mellitus

Abstract

Purpose.: To evaluate the effect of uncomplicated cataract phacoemulsification on macular and retinal nerve fiber layer (RNFL) thickness using two spectral-domain optical coherence tomography (OCT) instruments, Cirrus OCT and Spectralis OCT, in patients having non-insulin–dependent diabetes mellitus (NIDDM) without retinopathy, and to assess the reliability of the OCT measurements before and after cataract surgery.

Methods.: The study included 35 eyes of 35 patients having NIDDM without retinopathy (20 men and 15 women, mean age 69.8 years, range, 48–80 years) who underwent cataract phacoemulsification. One month before and 1 month after surgery, visual acuity and three repetitions of scans using the RNFL and macular analysis protocols of the Cirrus and Spectralis OCT instruments were performed. The differences between the two visits were analyzed by Student's t-test for paired samples. Repeatability of OCT measurements was evaluated by calculating the coefficients of variation for each of the parameters recorded and for each visit.

Results.: RNFL thicknesses provided by Cirrus and Spectralis OCT and macular measurements provided by Cirrus OCT differed significantly between the two visits. Macular thicknesses provided by Spectralis OCT before and after surgery were not significantly different. OCT repeatability was better after surgery with lower coefficients of variation for scans performed after surgical removal of the cataract.

Conclusions.: The presence of cataracts affects RNFL and macular measurements performed with OCT in NIDDM patients without retinopathy. The repeatability of the images significantly improved after cataract phacoemulsification.

Introduction

Diabetes mellitus type 2, non-insulin–dependent diabetes mellitus (NIDDM) or adult-onset diabetes, is a metabolic disorder characterized by high blood glucose levels in the context of insulin resistance and relative insulin deficiency, ¹ in contrast to diabetes mellitus type 1, in which there is an absolute insulin deficiency due to the destruction of pancreatic islet cells. ² The rates of diabetes have markedly increased over the last 50 years in parallel with obesity, which is the primary cause of type 2 diabetes in those with a genetic predisposition. In the developed world, and increasingly elsewhere, type 2 diabetes is the largest cause of nontraumatic blindness and kidney failure. ³  

Optical coherence tomography (OCT) is a well-recognized method of analyzing the in vivo retinal architecture. OCT is used for diagnosing and following retinal diseases, and is particularly useful and accurate for measuring retinal thickness. ^4,5 Several OCT studies evaluating the effects of cataract surgery in diabetic patients have been reported, but most of these evaluated the effect of anti-vascular endothelial growth factor or intravitreal corticoid injections during phacoemulsification to prevent macular edema in diabetic patients ⁶ or to reduce macular edema. ⁷  

NIDDM predisposes patients to macular edema, diabetic retinopathy, glaucoma, and optic nerve damage. ⁸ Noninvasive scanning using OCT yields valuable quantitative data related to the retinal nerve fiber layer (RNFL) and macular thickness. The clinical utility of OCT in NIDDM is well established. ^9,10  

Although OCT data generally have good repeatability, factors such as the presence of cataracts, the extent of pupillary dilation, and corneal dryness affect both the image quality and repeatability of subsequent OCT measurements. ^11–13 Cataracts may affect diagnostic OCT imaging quality, causing artifacts in the images. OCT is an important ophthalmic diagnostic tool, particularly in retinal diseases such as diabetic retinopathy. OCT imaging relies on near-infrared light. Similar to its effects in fundus photography and scanning laser ophthalmoscopy, a cataract is likely to increase light scattering and degrade OCT image quality. Some authors have reported variations in OCT measurements before and after uncomplicated cataract surgery, but these studies included only short series of healthy subjects and focused on a slight increase in macular thickness after surgery, ^14–16 and on the apparent increase in RNFL thickness due to the resulting improved image quality. ^17–20 To the best of our knowledge, no studies to date have compared the repeatability before and after cataract surgery using Fourier domain OCT in NIDDM patients. 

Repeatability and reproducibility are important for distinguishing normal subjects from patients with disease, as well as for following retinal or RNFL modifications over time for follow-up after treatment. ^21,22  

In the present study, we assessed the effects of cataracts on macular and RNFL thicknesses provided by two Spectralis domain OCT scans (Cirrus HD-OCT [Carl Zeiss Meditec, Inc., Dublin, CA] and Spectralis OCT [Heidelberg Engineering, Inc., Heidelberg, Germany]) in patients having NIDDM without retinopathy in the early postoperative period. We also evaluated the intraoperator reproducibility of OCT measurements in NIDDM patients before and after cataract extraction and intraocular lens implantation. 

Materials and Methods

This was an observational and prospective study. Thirty-five patients having a diagnosis of NIDDM and cataracts without retinopathy were enrolled in the study. The NIDDM diagnosis was based on the World Health Organization criteria. ²³ Because eye fixation is necessary to complete the ophthalmologic exploration, only NIDDM eyes with a best-corrected visual acuity of 0.1 or better on the Snellen scale were included. 

Exclusion criteria were the presence of significant refractive errors (>5 diopters of spherical equivalent refraction or 3 diopters of astigmatism); intraocular pressure of 21 mm Hg or higher; other causes of media opacification; systemic conditions that could affect the visual system; a history of ocular trauma or concomitant ocular diseases, including a previous history of optic neuritis; glaucoma; and laser therapy or ocular pathologies affecting the cornea, lens, or optic nerve. 

All procedures adhered to the tenets of the Declaration of Helsinki, and the experimental protocol was approved by the local ethics committee. All subjects provided informed consent to participate in the study and underwent a complete ophthalmologic evaluation, including pupillary and anterior segment examination; evaluation of type and degree of cataract according to the Lens Opacities Classification System III (LOCS III) system of nuclear color, nuclear opalescence, cortical, and posterior subcapsular cataract ²⁴ ; fundoscopic examination; assessment of best-corrected visual acuity relative to the Snellen scale; and three repetitions of scans using RNFL and macular cube 512 × 128 analysis protocols of the Cirrus OCT and the RNFL protocol of the Glaucoma application and Fast Retina protocol of the Retinal application of the Spectralis OCT. Glycosylated hemoglobin (HbA1) and plasma creatinine levels were determined. All measurements were performed at 1 month before and 1 month after cataract surgery. 

From 3 days before surgery to 7 days after surgery, patients were treated with moxifloxacin (5 mg/mL, Vigamox; Alcon Cusí, S.A., Barcelona, Spain) and diclofenac ophthalmic (Diclofenaco Lepori; Angelini Farmaceutica, S.A., Barcelona, Spain) eye drops; then during the 4-week postoperative period, 1% prednisolone acetate ophthalmic suspension (Pred Forte; Allergan, Inc., Irvine, CA) was applied with a progressive reduction of the dose. 

Surgery was performed under combined topical anesthesia with 0.1% tetracaine and 0.4% oxybuprocaine (Colircusí anestésico doble; Alcon Cusí, S.A.) at 10, 5, and 2 minutes before surgery. Surgery was performed by the same surgeon using small incisional phacoemulsification with an Alcon Infiniti Phaco System (Alcon Cusí, S.A.) with OZil available and implantation of endocapsular AcrySof intraocular lens with blue-light-filtering technology. ²⁵ Each eye was considered separately, and only one eye of each subject was included in the study. The indication for cataract surgery was reduced visual acuity below 0.6 on the Snellen scale or low enough to interfere with the patient's quality of life and daily activities, or when no satisfactory visual function could be obtained with glasses, contact lenses, or other optical aids. 

OCT tests were performed to obtain measurements of the macula and peripapillary RNFL using the Cirrus and Spectralis OCT devices, both of which were used in random order to prevent any effect of fatigue bias. The same experienced operator performed all scans. Between scan acquisitions, there was a time delay and subject position and focus were randomly disrupted, meaning that alignment parameters had to be newly adjusted at the start of each image acquisition. No manual correction was applied to the OCT output. An internal fixation target was used to provide the highest reproducibility. ²⁶ The quality of the scans was assessed prior to the analysis, and poor-quality scans were rejected. The Cirrus OCT determines the image quality using a signal strength measurement that combines the signal-to-noise ratio with the uniformity of the signal within a scan and is measured on a scale of 1 to 10, where 1 is categorized as poor image quality and 10 as excellent image quality. The Spectralis OCT uses a blue quality bar in the image to indicate the signal strength. The quality score range is 0 (poor quality) to 40 (excellent quality). Only two patients were excluded because a centered scan could not be acquired due to poor fixation (two other patients were substituted for these patients to maintain group size). Sixteen images with artifacts or missing parts or showing seemingly distorted anatomy were excluded. ²⁷ To obtain good-quality and centered images, five eyes required repeat scan acquisition using the Cirrus OCT, and three eyes required repeat scan acquisition using the Spectralis OCT. 

Following the recommended procedure for scan acquisition, the subject's pupil was first centered and focused in an iris-viewing camera on the system data acquisition screen, and then the system's line-scanning ophthalmoscope was used to optimize the view of the retina. The OCT scan was aligned to the proper depth and patient fixation, and system polarization was optimized to maximize the OCT signal. The software versions used were Cirrus OCT 6.0 and Spectralis OCT 5.4b. 

Three repetitions of optic disc cube 200 × 200 and macular cube 512 × 128 scans in each eye were performed using the Cirrus OCT. In each series of optic disc scans, the mean RNFL thickness and mean quadrant RNFL thickness (superior, inferior, temporal, and nasal) were analyzed. In macular scans, retinal thickness values were calculated for the nine areas corresponding to the Early Treatment Diabetic Retinopathy Study (ETDRS). ²⁸ The ETDRS areas include a central 1-mm circle, representing the foveal area, and inner and outer rings of 3- and 6-mm diameter, respectively. The inner and outer rings were divided into four quadrants: superior, nasal, inferior, and temporal. Central foveal thickness was also calculated. 

Three image acquisitions with the RNFL protocol of the classic Glaucoma application and Fast Retina protocol of the Retinal application using Spectralis OCT (Heidelberg Engineering, Inc.) with TruTrack eye-tracking technology were performed for all subjects. The mean number of scans to produce each circular B scan was nine. The RNFL protocol of the Spectralis generates a thickness map with mean total thickness, mean thickness in the four quadrants (superior, nasal, inferior, and temporal), and six sector thicknesses (superonasal, nasal, inferonasal, inferotemporal, temporal, and superotemporal in the clockwise direction for the right eye and counterclockwise for the left eye). The presence of defects in the RNFL was determined by comparing the measurements from each patient with the normative database of each instrument. The retinal thickness values were similar to those using Cirrus OCT: The nine areas corresponding to the ETDRS study were calculated. 

Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS 19.0; SPSS, Inc., Chicago, IL). The Kolmogorov-Smirnov test was used to assess sample distribution. RNFL and macular parameters were compared before and after cataract surgery using Student's t-test for paired data. P values less than 0.05 were considered to indicate statistically significant differences. 

For each OCT parameter, the coefficient of variation (COV) was calculated as the standard deviation divided by the mean of the measurement value and expressed as a percentage. Most investigators consider that devices with a COV less than 10% have high repeatability, while a COV less than 5% indicates very high repeatability. ²² Bland-Altman plots were used to assess agreement. 

Spearman correlation analysis revealed linear agreement between the degree of cataract according to LOCS III and the change in thickness measurements provided by the OCT before and after surgery. 

Results

Thirty-five eyes from 35 subjects (20 men and 15 women, mean age 69.8 years, range, 48–80 years) were examined. Best-corrected visual acuity was 0.31 ± 0.18 before cataract surgery and 0.84 ± 0.19 after surgery (Table 1). Mean NIDDM duration in the patients was 9.6 ± 2.1 years, and mean age at NIDDM diagnosis was 63.9 ± 3.9 years. Of the 35 patients, 18 presented with intermediate cataracts (7 nuclear cataracts, 7 nuclear-cortical cataracts, and 4 posterior subcapsular cataracts) and 17 presented with advanced cataracts (6 nuclear cataracts, 7 nuclear-cortical cataracts, 2 posterior subcapsular cataracts, and 2 nuclear-cortical-posterior subcapsular cataracts). 

Table 1

 

Epidemiologic Characteristics of the 35 Subjects With NIDDM and Cataract Included in the Study and Comparison of Parameters Between the Visit Before Surgery and the Visit 1 Month After Cataract Surgery

	Before Surgery	After Surgery	P
BCVA, Snellen scale	0.31 (0.1)	0.84 (0.1)	<0.001
Glycosylated hemoglobin, %	10.7 (1.0)	10.6 (1.1)	0.326
Plasma creatinine levels, μmol/L	85.7 (9.1)	86.1 (8.7)	0.128
Systolic blood pressure, mm Hg	142.8 (8.1)	142.2 (6.7)	0.664
Diastolic blood pressure, mm Hg	86.6 (7.4)	85.9 (7.0)	0.519
Intraocular pressure, mm Hg	15.26 (1.32)	15.36 (1.44)	0.202

 Values are given as mean (standard deviation).

RNFL Thickness Comparison Before and After Cataract Surgery

We compared RNFL parameters obtained 1 month before and 1 month after cataract surgery using the Cirrus and Spectralis OCT devices. Mean RNFL thickness measurements, based on three individual scans, were used for the analysis (Table 2). The RNFL thickness measured by both devices differed significantly between before and after surgery. With the Cirrus OCT, statistical differences were observed in mean thickness and superior and nasal quadrants (P < 0.05), with the greatest difference in the superior quadrant (2.3 μm thicker after surgery, P = 0.008). With the Glaucoma application of the Spectralis OCT, the superior, inferotemporal, and temporal RNFL sector thicknesses and mean total RNFL thickness were significantly greater after surgery. The largest difference was in the inferotemporal sector (3.6 μm thicker after surgery, P = 0.017). 

Table 2

 

Mean and Standard Deviation of RNFL Thickness Measurements of 35 Patients With NIDDM Using Cirrus and Spectralis OCT (Glaucoma Application) Before and After Cataract Surgery and Statistical Significance

	Before Surgery	After Surgery	P
Cirrus RNFL OCT parameters
Mean thickness	80.7 ± 12.5	82.2 ± 16.3	0.006
Superior thickness	102.4 ± 15.9	104.7 ± 16.8	0.008
Nasal thickness	66.7 ± 14.8	67.9 ± 15.7	0.023
Inferior thickness	103.2 ± 14.9	103.8 ± 15.8	0.345
Temporal thickness	56.8 ± 12.7	57.1 ± 14.6	0.108
Glaucoma application of Spectralis RNFL OCT parameters
Mean thickness	90.2 ± 15.8	92.8 ± 14.6	0.027
Superior thickness	106.8 ± 10.8	109.2 ± 13.9	<0.001
Inferior thickness	112.9 ± 13.5	113.4 ± 11.6	0.257
Superonasal thickness	93.7 ± 12.8	95.5 ± 15.8	0.455
Nasal thickness	67.6 ± 14.9	67.4 ± 13.3	0.199
Inferonasal thickness	100.9 ± 16.5	101.2 ± 14.9	0.440
Inferotemporal thickness	121.7 ± 17.9	125.3 ± 16.0	0.017
Temporal thickness	73.2 ± 16.8	76.0 ± 14.1	0.045
Superotemporal thickness	118.5 ± 17.1	120.1 ± 13.9	0.059

 All measurements are in micrometers.

Figure 1 shows the RNFL thickness differences before and after surgery for the mean total thickness and mean thickness of each of the four quadrants measured by the Cirrus OCT (Fig. 1A) and the Glaucoma application of the Spectralis OCT (Fig. 1B). In all measurements, the RNFL thickness was slightly increased after surgery. 

Figure 1

 

Representation of retinal nerve fiber layer thicknesses using Cirrus (A) and Glaucoma application of Spectralis optical coherence tomography (B) in 35 eyes of 35 patients with non-insulin–dependent diabetes mellitus and cataract before and after surgery. Measurements are higher after surgery. RNFL, retinal nerve fiber layer; OCT, optical coherence tomography. All measurements are in micrometers (μm).

Macular Thickness Comparison Before and After Cataract Surgery

We compared all structural macular parameters obtained by the Cirrus and Spectralis OCTs in the two visits, 1 month before and 1 month after cataract surgery (Table 3). Mean thickness measurements, based on three individual scans, were used for the analysis. All macular parameters were significantly increased after surgery based on measurements with the Cirrus OCT (Fig. 2A). The largest difference was in the superior outer thickness (34.3 μm higher after surgery, P = 0.005). Macular thickness measurements made with the Spectralis OCT, however, were not significantly different after surgery (Fig. 2B). 

Figure 2

 

Representation of nine Early Treatment Diabetic Retinopathy Study (ETDRS) thicknesses using Cirrus (A) and Spectralis optical coherence tomography (B) in 35 eyes of 35 patients with non-insulin–dependent diabetes mellitus and cataract before and after surgery. Measurements are higher after surgery. RNFL, retinal nerve fiber layer; OCT, optical coherence tomography. All measurements are in micrometers (μm).

Table 3

 

Mean and Standard Deviation of Macular Structural Measurements of 35 Patients With NIDDM Using Cirrus and Spectralis OCT Before and After Cataract Surgery and Statistical Significance (P) of Comparison Between Visits

	Before Surgery	After Surgery	P
Cirrus macular OCT parameters
Fovea	230.1 ± 15.9	258.0 ± 15.0	<0.001
Superior inner macula	303.7 ± 14.6	321.2 ± 16.9	0.004
Nasal inner macula	300.1 ± 18.3	319.7 ± 17.1	0.022
Inferior inner macula	291.1 ± 18.9	301.7 ± 18.6	0.038
Temporal inner macula	271.9 ± 17.3	295.7 ± 16.2	0.001
Superior outer macula	244.7 ± 15.5	279.0 ± 14.8	0.005
Nasal outer macula	265.8 ± 14.8	295.1 ± 15.3	<0.001
Inferior outer macula	261.5 ± 15.9	274.4 ± 14.4	0.002
Temporal outer macula	258.0 ± 14.1	266.6 ± 17.8	0.009
Spectralis macular OCT parameters
Fovea	309.1 ± 13.4	310.6 ± 14.6	0.222
Superior inner macula	356.5 ± 15.8	355.2 ± 13.7	0.502
Nasal inner macula	345.6 ± 16.0	344.2 ± 13.3	0.211
Inferior inner macula	324.7 ± 18.2	325.0 ± 14.9	0.598
Temporal inner macula	341.2 ± 17.8	343.5 ± 18.9	0.519
Superior outer macula	300.2 ± 17.0	301.9 ± 19.1	0.626
Nasal outer macula	299.3 ± 16.7	296.6 ± 16.3	0.308
Inferior outer macula	312.7 ± 15.8	309.8 ± 15.0	0.447
Temporal outer macula	289.1 ± 15.1	290.4 ± 14.2	0.328

 All measurements are in micrometers.

Repeatability of Cirrus and Spectralis OCT Before and After Cataract Surgery

RNFL and macular thickness measurements had lower COVs after surgery (Table 4, Figs. 3, 4). RNFL measurements obtained with the Cirrus OCT were highly reproducible after surgery for all quadrants and sectors: Mean COV before surgery was 9.11 ± 5.44%, and after surgery it was 5.99 ± 4.11% (P < 0.001). Mean thickness after surgery had the lowest variability (COV = 2.33%). Mean thickness and superior, inferior, and temporal quadrant COV values were significantly different between before and after cataract surgery (P = 0.004, P = 0.038, P = 0.045, and P = 0.012, respectively; Table 4). 

Figure 3

 

Graph of the agreement in retinal nerve fiber layer mean thickness using Cirrus ([A], before surgery; [B], after surgery) and Glaucoma application of Spectralis optical coherence tomography ([C], before surgery; [D], after surgery) in 35 eyes of 35 patients with non-insulin–dependent diabetes mellitus and cataract. The Bland-Altman plots represent the difference (mean thickness measurement 1–2) against the mean of the three measurements of mean thickness. RNFL, retinal nerve fiber layer; OCT, optical coherence tomography.

Figure 4

 

Graph of the agreement in macular mean thickness using Cirrus ([A], before surgery; [B], after surgery) and Spectralis optical coherence tomography ([C], before surgery; [D], after surgery) in 35 eyes of 35 patients with non-insulin–dependent diabetes mellitus and cataract. The difference (mean thickness measurement 1–2) is represented against the mean of the three measurements of mean thickness. The Bland-Altman plots represent the difference (mean thickness measurement 1–2) against the mean of the three measurements of mean thickness. RNFL, retinal nerve fiber layer; OCT, optical coherence tomography.

Table 4

 

Coefficients of Variation for Repeated RNFL Measurements and Repeated Macular Structural Measurements and Statistical Significance With Cirrus and Spectralis OCT in 35 Patients With NIDDM Before and After Surgery

	COV Before Surgery	COV After Surgery	P
RNFL thickness measurements
Cirrus high-definition OCT parameters
Mean thickness	7.01	2.33	0.004
Superior quadrant	8.34	5.71	0.038
Nasal quadrant	10.58	8.90	0.078
Inferior quadrant	10.12	7.55	0.045
Temporal quadrant	9.51	5.48	0.012
Spectralis OCT parameters
Mean thickness	2.59	2.01	0.038
Superior thickness	3.66	3.11	0.059
Inferior thickness	3.01	2.14	0.043
Superonasal thickness	5.37	4.71	0.344
Nasal thickness	4.88	4.25	0.767
Inferonasal thickness	5.56	4.76	0.412
Inferotemporal thickness	4.00	3.78	0.500
Temporal thickness	4.12	3.70	0.672
Superotemporal thickness	3.81	3.62	0.381
Macular structural measurements
Cirrus high-definition OCT parameters
Fovea	8.12	5.45	<0.001
Superior inner macula	8.14	4.05	<0.001
Nasal inner macula	10.2	6.33	0.015
Inferior inner macula	9.01	7.98	0.241
Temporal inner macula	8.80	3.61	0.035
Superior outer macula	9.99	4.91	0.041
Nasal outer macula	6.12	5.12	0.167
Inferior outer macula	6.29	4.76	0.044
Temporal outer macula	5.87	5.44	0.078
Spectralis OCT parameters
Fovea	6.55	2.41	0.028
Superior inner macula	5.78	2.88	0.019
Nasal inner macula	5.01	4.56	0.057
Inferior inner macula	4.66	1.56	0.030
Temporal inner macula	5.77	4.39	0.339
Superior outer macula	6.01	5.47	0.731
Nasal outer macula	6.99	5.61	0.798
Inferior outer macula	7.37	4.22	0.043
Temporal outer macula	5.74	4.10	0.642

 Coefficients of variation after surgery are given in percentages.

Measurements performed using the Glaucoma application of the Spectralis OCT had better repeatability after surgery. COVs differed significantly between before and after surgery only for mean thickness (P = 0.038) and for the inferior quadrant (P = 0.043). Mean COV using this application before surgery was 4.11 ± 2.89%, and after surgery it was 3.56 ± 2.03% (P = 0.078). Mean thickness after surgery had the lowest variability (COV = 2.01%). Figure 3 shows Bland-Altman plots of RNFL mean thickness repeatability before and after cataract surgery using the Cirrus OCT and the Glaucoma application of the Spectralis OCT. 

The COV of the macular thickness measurements was lower after surgery with both devices. The COVs differed significantly before and after surgery in the fovea: superior inner, nasal inner, temporal inner, superior outer, and inferior outer macular thicknesses using the Cirrus OCT, and superior inner, inferior inner, and inferior outer macular thicknesses using the Spectralis OCT (Table 4). The temporal inner macular thickness after surgery had the lowest variability using the Cirrus OCT (COV = 3.61%), and the inferior inner macular thickness after surgery had the lowest variability using the Spectralis OCT (COV = 1.56%). Figure 4 shows Bland-Altman plots of foveal thickness repeatability before and after cataract surgery using the Cirrus and Spectralis OCTs. 

The type and degree of cataract were not significantly associated with thickness changes registered after surgery. 

Discussion

Although OCT is widely used by ophthalmologists to diagnose and monitor patients, the effect of cataracts on OCT measurements has scarcely been investigated. Several studies suggest that cataracts reduce image quality, although this effect seems to depend on the type of cataract, with nuclear cataracts having less effect than cortical and posterior cataracts on OCT image quality. ^11–13 Some studies have demonstrated that RNFL and macular thicknesses increase after surgery in healthy subjects. ^{14,17,19,26,29,30} Most previous studies were performed using healthy subjects or patients with glaucoma. ^14,18,19,29 to compare Stratus and Cirrus OCT images before and after cataract surgery, but the Spectralis OCT has not been evaluated. ^14,19,28  

To our knowledge, there are no published studies of NIDDM patients in which the effect of cataracts on RNFL parameters were examined or OCT measurement repeatability was compared. Some authors compared OCT evaluation of the eyes of diabetic patients and the eyes of healthy controls, and reported a greater foveal thickness and a higher incidence of macular edema after cataract surgery in diabetic patients, ³¹ especially patients with retinopathy, ^32,33 but also in patients without retinopathy. ^34,35  

Our study evaluated changes in RNFL and macular thicknesses after uncomplicated cataract phacoemulsification in patients diagnosed with NIDDM, and compared repeatability before and after surgery using the Cirrus and Spectralis OCT devices in these patients. In the present study, postoperative OCT scans showed a considerable increase in RNFL and macular thicknesses and improved repeatability after cataract removal in NIDDM patients. The focus of the present study was to examine patients with a definitive diagnosis of NIDDM and compare the effects of cataracts on measurements performed using two of the most modern and extended spectral-domain OCT devices in clinical practice, Cirrus and Spectralis. 

OCT instruments have recently become very useful for analyzing optic disc and macular disorders. ^36,37 The presence of cataracts causes light scattering on OCT acquisition, which produces distortions that affect the OCT quality and measurements. In the present study, the OCT measurements were statistically different before and after cataract phacoemulsification, and thus the effects of cataracts in NIDDM patients should be considered when following up changes in these patients, because an increase or decrease in RNFL or retinal thickness may be due to the presence or absence of the cataract rather than an actual pathologic change associated with NIDDM. In addition, several studies have demonstrated a higher risk of macular edema in NIDDM patients compared with healthy subjects, even in patients without signs of diabetic retinopathy in funduscopy before and after surgery. ^34,38,39 Biró and Balla ³⁹ evaluated changes in macular thickness in NIDDM patients and healthy subjects and found no significant change in the thickness values 1 day after surgery, but detected a significant increase on postoperative days 7, 30, and 60. 

We found a significant thickness increase in the nine ETDRS areas using the Cirrus OCT, but not when using the Spectralis OCT. This observation is particularly interesting in NIDDM patients whose progressive macular thinning is monitored by OCT. In these cases, if the follow-up after cataract surgery is performed using a Cirrus OCT, the ophthalmologist must consider that the observed changes could be due to the variability of the measurements associated with cataract removal. With use of the Spectralis OCT, however, changes in macular thickness detected after surgery should be due solely to NIDDM progression. The association between the LOCS III scale and thickness changes was not statistically significant, but the size of the cataract groups was small. Studies with larger groups of patients may reveal a correlation with the type or degree of cataract. 

The findings of our study indicate that the Cirrus and Spectralis OCTs have better repeatability after cataract removal. The clinical implication of these results is important and indicates that measurements obtained with OCT in patients with NIDDM and cataract should be interpreted cautiously. Changes observed in the macular or RNFL thicknesses may be due to the variability associated with the presence of a cataract and not an actual change in the NIDDM pathology; but NIDDM patients may present with macular edema after surgery, which could affect macular thickness measurements provided by OCT. Furthermore, our results reveal some parameters whose variability improved more than others after cataract extraction: The repeatability of the mean, superior, inferior, and temporal quadrant RNFL thicknesses and six of the nine retinal ETDRS areas was significantly improved after surgery using Cirrus OCT. It is noteworthy that in the case of a discrepancy in the measurements, the most reliable parameters for the ophthalmologist (because they were the least variable) are the RNFL mean thickness, the temporal inner macular thickness with Cirrus OCT, and the inferior inner area thickness with Spectralis OCT after surgery. The mean RNFL thickness is reported to be less variable in other studies evaluating both healthy and pathologic eyes. ³⁶  

Our findings revealed higher variability of OCT measurements compared with that in reproducibility studies evaluating healthy subjects. The reason for the lower reproducibility in our study might be that all the subjects included suffered retinal pathology. NIDDM causes macular edema, glaucoma, and ganglion cell atrophy. ^8–10 All these alterations generate changes in OCT measurements, so RNFL and retina thicknesses are more variable in pathologic eyes than in healthy subjects. 

Previous studies have described nasal RNFL thicknesses as the most variable of the OCT parameters and mean RNFL thickness as the least variable OCT parameter. ^22,36,40 The present results confirmed these findings in both visits (before and after surgery) and with both OCT devices (Cirrus and Spectralis). This variability in RNFL nasal sectors could be explained in terms of the greater difficulty of using the measurement algorithm to calculate RNFL thickness in the nasal quadrant. 

Before surgery, the Spectralis OCT (with TruTrack technology) exhibited lower variability than the Cirrus OCT device. The TruTrack technology seems to improve measurement repeatability because this system locks onto and follows the patient's retina and optic nerve during the scanning independent of eye fixation, which improves repeatability. This might explain the decreased differences in the COVs observed before and after cataract surgery with use of the Spectralis OCT. For this reason, we advise using the Spectralis OCT in NIDDM patients with cataracts. After cataract removal, however, both devices have very good repeatability. The scan acquisition speed is faster with Spectralis OCT than with Cirrus OCT, which may be the cause of the lower variability detected with the Spectralis device. 

Longer prospective studies using spectral-domain OCT to analyze the changes in RNFL and macular thickness measurements after cataract surgery in patients with NIDDM or other ophthalmologic pathologies are needed. 

In conclusion, one must take the effect of cataracts on OCT measurements into account when interpreting OCT scans in NIDDM patients. Even in the presence of cataracts, OCT scans of individual patients remain reliable for clinical interpretation of NIDDM pathology, although it should be noted that RNFL and macular thickness values might be lower due to the presence of the cataract. 

Acknowledgments

Disclosure: E. Garcia-Martin, None; J. Fernandez, None; L. Gil-Arribas, None; V. Polo, None; J.M. Larrosa, None; S. Otin, None; I. Fuertes, None; L. Pablo, None