Wickham Louisa, Edmunds Beth, Murdoch Ian E
Moorfields Eye Hospital, London, United Kingdom.
Ophthalmology. 2005 Feb;112(2):225-8. doi: 10.1016/j.ophtha.2004.08.020.
To evaluate the measurement of central corneal thickness (CCT) in a cohort of glaucoma patients over a 3-month period. Measurements were then applied to the criteria described in the Ocular Hypertension Treatment Study (OHTS).
Cohort study.
Fifty-one patients were recruited from a glaucoma clinic at Moorfields Eye Hospital. Central corneal thickness was measured using an ultrasonic handheld pachymeter by a trained observer. Patients' CCTs were measured at 2 consecutive clinic visits.
Mean central corneal thickness.
The readings showed clear fluctuation over the 3-month period, with a mean difference in corneal thickness of 9.6+/-26.9 microm in the right eye and 19.0+/-29.2 microm in the left eye. In addition, there was a systematic bias towards increased corneal thickness being recorded at the second reading in both eyes. This reached statistical significance in both the right eye (P = 0.02) and the left eye (P = 0.0003). The criteria used to categorize the risk of patients developing glaucoma in the OHTS were then applied to these results. On the basis of the second reading, 32% of eyes required recategorization in both the right and left eyes.
Measurements of CCT taken within a clinical setting by a trained observer may show significant variability. For CCT to become a valuable addition to the assessment of glaucoma suspects, more than one reading may be required. Failure to do so may result in misclassification and, thus, an inaccurate assignment of risk.
评估一组青光眼患者在3个月期间中央角膜厚度(CCT)的测量情况。然后将测量结果应用于高眼压治疗研究(OHTS)中描述的标准。
队列研究。
从摩尔菲尔德眼科医院的青光眼诊所招募了51名患者。由一名经过培训的观察者使用超声手持测厚仪测量中央角膜厚度。在连续两次门诊就诊时测量患者的CCT。
平均中央角膜厚度。
读数在3个月期间显示出明显波动,右眼角膜厚度的平均差异为9.6±26.9微米,左眼为19.0±29.2微米。此外,双眼在第二次读数时记录的角膜厚度均有系统性增加偏差。这在右眼(P = 0.02)和左眼(P = 0.0003)中均达到统计学显著性。然后将OHTS中用于对患者发生青光眼风险进行分类的标准应用于这些结果。根据第二次读数,右眼和左眼均有32%的眼睛需要重新分类。
由经过培训的观察者在临床环境中进行的CCT测量可能显示出显著的变异性。为使CCT成为青光眼疑似患者评估中有价值的补充指标,可能需要进行不止一次读数。否则可能导致错误分类,从而对风险的分配不准确。