- Email: [email protected]
Intraocular pressure difference in Goldmann applanation tonometry versus Perkins hand-held applanation tonometry in overweight patients
Intraocular Pressure Difference in Goldmann Applanation Tonometry versus Perkins Hand-held Applanation Tonometry in Overweight Patients Mario Gonzaga dos Santos, MD,1 Stefan Makk, MD,1 Andrea Berghold, PhD,2 Martin Eckhardt, MD,1 Anton Haas, MD1 Objective: To analyze the increase in intraocular pressure (IOP) caused by anatomic and physiologic factors in overweight patients when using Goldmann applanation tonometry. Design: A prospective cohort study. Participants: Seventy average-weight individuals who had no difficulties with IOP measurements at the slit lamp and 12 obese patients with suspected glaucoma who could position the head at the slit lamp only with great effort participated. Intervention: The authors compared IOP values between slit-lamp-mounted Goldmann applanation tonometry and Perkins hand-held tonometry. Main Outcome Measure: The difference in Goldmann and Perkins IOP measurements was examined. Results: In the group of obese patients, the mean IOP was 20.9 ⫾ 2.28 mmHg (mean ⫾ standard deviation; range, 18 –26 mmHg) for the right eye and 21.4 ⫾ 3.16 mmHg (range, 16 –28 mmHg) for the left eye when determined by Goldmann tonometry and 16.3 ⫾ 2.39 mmHg (range, 13–20 mmHg) for the right eye and 16.3 ⫾ 2.42 (range, 11–19 mmHg) for the left eye when determined by Perkins tonometry. The mean decrease was 4.5 ⫾ 1.3 mmHg (range, 3–7 mmHg) for the right eye and 4.9 ⫾ 1.9 mmHg (range, 2–9 mmHg) for the left eye. In the control group, the mean difference between the two types of tonometers for the right eye was 0.34 ⫾ 0.69 mmHg and for the left eye was 0.33 ⫾ 0.82 mmHg. Patients who had a falsely elevated IOP on Goldmann tonometry had an average body mass index of 34 ⫾ 3.82 (range, 28.5– 41.9); most were female (5:1 ratio). Conclusion: The authors believe simultaneous breath-holding and thorax compression, with subsequent increase in venous pressure, may be a causative factor for transitory elevations of IOP. Perkins tonometry in obese patients may help avoid a false diagnosis of glaucoma caused by transitory elevations in IOP. Ophthalmology 1998;105:2260 –2263 Although visual field defects are widely regarded as the most important and definite sign of glaucoma, elevation of intraocular pressure (IOP) is still an important feature in the diagnosis and treatment of glaucoma. Currently, the most common way to assess IOP is with the slit-lamp-mounted Goldmann applanation tonometer. Although the Goldmann-type tonometer has considerable reproducibility and accuracy, we should be aware of sources of error using this type of tonometer. Any circumstances that cause a reduction in the fluorescence of the precorneal tear film will cause an underestimation of IOP.1 Corneal Originally received: July 31, 1997. Revision accepted: July 6, 1998. Manuscript no. 97428. 1 Department of Ophthalmology, Karl-Franzens University of Graz, Graz, Austria. 2 Institute of Medical Informatics, Statistics and Documentation, KarlFranzens University of Graz, Graz, Austria. Reprint requests to Mario Gonzaga dos Santos, MD, Department of Ophthalmology, Karl-Franzens University, Auenbruggerplatz 4, A-8036 Graz, Austria.
2260
epithelial abnormalities, corneal thickness, and astigmatism influence the results of IOP measurements with the Goldmann applanation tonometer.2 Contact between the applanating prism and the eyelashes induces an increase in the apparent IOP. Repeated tonometry induces a decline in IOP.3 Changes in blood circulation, such as increased venous pressure from a Valsalva maneuver, increase IOP.4 Breath-holding may also be associated with the Valsalva maneuver, causing an increase of IOP, and may be an additional source of significant error in applanation tonometry. This phenomenon can be observed, especially in overweight patients who have difficulties in taking up the correct position at the slit lamp during IOP measurements because of a short, stocky build, short neck, and a large chest and abdomen. Epstein5 suggested using the Perkins hand-held tonometer on obese patients to avoid this increase in IOP induced by slit-lamp tonometry. In our glaucoma service, we became aware of a series of obese patients who had persistently elevated IOP for years but no visual field defects and no optic disc pathology. All had been treated for glaucoma for years. Using the Perkins hand-held tonometer,
dos Santos et al 䡠 Tonometry in Overweight Patients the IOP values were within the normal range. In this study, we investigated the extent of the IOP increase caused by anatomic and physiologic factors when using Goldmann applanation tonometry in obese patients.
Patients and Methods We prospectively investigated anatomic and physiologic factors influencing IOP differences between slit-lamp-mounted Goldmann applanation tonometry and Perkins hand-held tonometry in patients with anatomic difficulties in positioning the head at the slit lamp for IOP measurements. Patients were divided into two groups. Group 1 consisted of 12 overweight patients (10 female and 2 male) with positioning difficulties. The mean age was 67.7 ⫾ 8 years (mean ⫾ standard deviation; range, 53–78 years). The patients in this group had a body habitus characterized by a stocky build, a broad face, short neck, and a large chest and abdomen and could hold the head at the slit lamp only with great effort. Nine patients in this group had been treated for primary open-angle glaucoma for 2 to 19 years. Three patients were suspected of having glaucoma. All 12 of these patients had a normal cup– disc ratio and normal visual fields. Three weeks before beginning the study, the patients stopped using all glaucoma medication. The IOP values of these patients were compared to those of subjects of a control group (group 2), which consisted of 70 subjects who had no difficulty in positioning the head at the slit lamp. Forty-five of these subjects had no ocular pathology. Five patients had newly diagnosed glaucoma without any treatment, and 20 patients had primary open-angle glaucoma without filtering surgery. The average age for the control group was 64.9 ⫾ 12 years (range, 33– 89 years). Clinical signs such as the compression of the patient against the slit lamp, effort required to hold the position, breath-holding, and hyperemia of the face due to venous congestion were considered. “Borderline” patients, in which there was doubt as to whether they were having difficulty in positioning at the slit lamp, were excluded from both groups. Body mass index (BMI) was used as a method to define obesity but was not used to divide the participants into the two groups. BMI was calculated with the formula: weight (kg)/height (m2). A BMI between 20 and 25 is considered to be within the normal range. Obesity is taken to start at a BMI of 30. The group of overweight patients had an average BMI of 34.9 ⫾ 3.82, corresponding to an average weight of 91.8 kg and a height of 162 cm. The control group had an average BMI of 24.9 ⫾ 3.45, corresponding to an average weight of 70.1 kg and a height of 169 cm. Intraocular pressure was measured with the Goldmann applanation tonometer on a slit-lamp biomicroscope (BQ 900; Haag– Streit AG, Bern, Switzerland; slit-lamp table: Block, Dortmund, Table 1. Intraocular Pressure Measurements with the Perkins Hand-held Tonometer and the Goldmann Slit-lamp Mounted Tonometer of the Overweight Patients (n ⫽ 12) Goldmann
Mean SD Minimum Maximum Range
Perkins
RE
LE
RE
LE
20.6 2.28 18.0 26.0 8
21.0 3.16 16.0 28.0 12
16.3 2.39 13.0 20.0 7
16.3 2.42 11.0 19.0 8
RE ⫽ right eye; LE ⫽ left eye; SD ⫽ standard deviation.
Table 2. Intraocular Pressure Measurements of the Perkins Hand-held Tonometer and the Goldmann Slit-lamp Mounted Tonometer of the Control Subjects (n ⫽ 70) Goldmann
Mean SD Minimum Maximum Range
Perkins
RE
LE
RE
LE
16.8 3.71 8 26.0 21
18.0 4.2 10.0 32.0 22
16.4 3.72 8.0 26.0 21
17.7 4.33 10.0 32.0 22
RE ⫽ right eye; LE ⫽ left eye; SD ⫽ standard deviation.
Germany) using the method of the Advanced Glaucoma Intervention Study, and with the Perkins hand-held applanation tonometer (Clement Clarke International Ltd, Harlow, England).6,7 Perkins and Goldmann tonometers were calibrated before the beginning of the study. The IOP measurements were performed starting either with Goldmann or with Perkins tonometry and then reversing the sequence after an interval of 10 minutes. All patients measured with the Perkins tonometer were measured in the sitting position. By keeping the test conditions constant, we tried to minimize physiologic changes in IOP other than changes due to the efforts of the overweight patients to position themselves at the slit lamp. No respiratory maneuvers or special movements were made to try to modify or normalize the IOP. The Wilcoxon rank–sum test was used to analyze differences. A P value less than 0.001 was considered statistically significant.
Results In the group of obese patients, the mean IOP was 20.6 ⫾ 2.28 mmHg (range, 18 –26 mmHg) for the right eye and 21.0 ⫾ 3.16 mmHg (range, 16 –28 mmHg) for the left eye as determined by Goldmann tonometry, and 16.3 ⫾ 2.39 mmHg (range, 13–20 mmHg) for the right eye and 16.3 ⫾ 2.42 mmHg (range, 11–19 mmHg) for the left eye when determined by Perkins tonometry. The mean IOP with the Goldmann tonometer of the control group was 16.8 ⫾ 3.71 mmHg (range, 8 –26 mmHg) for the right eye and 18.0 ⫾ 4.2 mmHg (range, 10 –27 mmHg) for the left eye. Very similar IOP values were obtained in this group with the Perkins tonometer with 16.4 ⫾ 3.72 mmHg (range, 8 –26 mmHg) for the right eye and 17.7 ⫾ 4.33 mmHg (range, 10 –27 mmHg) for the left eye (Tables 1 and 2). When analyzing the mean differences between the two tonometers in the obese patients, we found a significant IOP decrease of 4.5 ⫾ 1.3 mmHg (range, 3–7 mmHg) for the right eye and 4.9 ⫾ 1.9 mmHg (range, 2–9 mmHg) for the left eye when measured with the Perkins tonometer. In the control group, the mean difference between the two types of tonometers for the right eye was 0.34 ⫾ 0.69 mmHg and for the left eye was 0.33 ⫾ 0.82 mmHg, which was not statistically significant. When comparing both groups, these values were statistically significant for both eyes (P ⬍ 0.001). In the group of obese patients, the mean BMI of 34.9 ⫾ 3.82 (range, 28.5– 41.9) was significantly lower (P ⬍ 0.001) compared to the control group (24.9 ⫾ 3.45; range, 20.1–33.6). No statistically significant difference was observed regarding the age between the two groups. There was a 5:1 ratio of women to men in the group of obese patients, whereas the control group had an equal gender distribution.
2261
Ophthalmology Volume 105, Number 12, December 1998
Discussion Although slit-lamp-mounted Goldmann applanation tonometry is the most common way to measure IOP and is characterized by good accuracy and reproducibility, ophthalmologists should be aware of sources of underestimation or overestimation to avoid false diagnosis and unnecessary treatment of glaucoma.8 This study describes a series of overweight patients in our glaucoma service who had falsely been diagnosed as having glaucoma. The IOP was significantly higher in these patients when measured with Goldmann tonometry compared to Perkins hand-held tonometry. Anatomic and mechanical difficulties in putting the chin in the right position at the slit lamp cause muscular contraction, breath-holding, and anxiety, with effects on the heart and circulatory system. Pushing thorax and abdomen against the slit lamp and table while breath-holding works like a Valsalva maneuver. Valsalva-like maneuvers frequently are encountered in many daily activities like lifting heavy loads, defecation, playing wind instruments, coughing, and vomiting.9 The Valsalva maneuver, a procedure in which the intrathoracic pressure is raised by an expiratory effort, produces a variety of hemodynamic responses. In patients with a normal cardiac condition, the central venous pressure increases and the systolic and diastolic pressures decrease. Elevation of central venous pressure acts as a back pressure on the systemic circulation and opposes the return of blood from the peripheral blood vessels into the heart. Positive or negative pressures caused by normal or abnormal respiration frequently alter the central venous pressure as much as 4 to 5 mmHg, and occasionally as much as 20 to 30 mmHg. Elevation in venous pressure communicates through the jugular, orbital, and vortex veins to the choroid and can therefore increase the IOP in two different ways: 1. The elevated pressure in the orbital veins interferes with the outflow of aqueous from the anterior chamber. 2. The elevated pressure can cause venous stasis, with vascular engorgement and increase in the volume of the choroidal veins. During the Valsalva maneuver, the acute alteration of the IOP is more likely to be caused by venous stasis with a rise in intraocular volume. Glaucoma in superior vena caval obstruction syndrome is a good clinical example that directly demonstrates the relationship between elevated venous pressure and the IOP.10 The importance of respiratory pressure as a cause of increase in thoracic pressure and subsequent elevation of venous and ocular pressures was shown by Rosen and Johnston,11 who described ocular pressure patterns in the Valsalva maneuver. The force of the expiratory effort was found to have a considerable influence on the magnitude of the ocular pressure response, showing an almost linear increase in IOP with increasing force of expiration. Several patients showed IOP increases up 20 mmHg. Bain and Maurice12 produced an elevation of venous pressure by inflating a sphygmomanometer cuff around the
2262
neck to 40 mmHg, which almost doubled the IOP. Constrictive clothing around the neck may acutely increase the IOP as the neck is extended when positioning the head at the slit lamp.13 Repeated tonometer contact induces a reduction of the IOP.14 –16 The extent of the reduction depends on the number of contacts and on the interval of time between the measurements. To exclude this possible factor of error in our study, IOP measurements were performed in the eyes in a reverse sequence after an interval of 10 minutes. The difference between the decrease of IOP from Goldmann to Perkins tonometry and the rise in the values when the IOP was measured in reverse order was statistically insignificant in both groups. Perkins hand-held tonometry as a method of IOP measurement has been assessed previously.8,17,18 In the study of Whitty,16 all readings taken with the hand applanator were found to be lower than those taken with the Haag–Streit tonometer. A difference of 1.0 mmHg and 1.5 mmHg was thought to be acceptable. In our control group, the mean difference between Perkins and Goldmann tonometry was 0.34 ⫾ 0.69 mmHg for the right eye and 0.33 ⫾ 0.83 mmHg for the left eye, thus confirming good reproducibility. In comparison to the Goldmann tonometer, the Perkins tonometer is more difficult to use. The illumination is weaker, and it is harder to see the fluorescein half-rings compared to the Goldmann tonometer. Despite the relative difficulty for the patients to maintain a steady head posture without a support and for the user to hold the apparatus with a steady hand and wrist, use of the Perkins tonometer can be learned after a short time of practice. The tendency to take slightly lower readings in the case of premature cessation of the tonometry was taken into account. The standard deviation between Goldmann and Perkins tonometers in the control group did not exceed ⫾0.82 mmHg and confirms the Perkins tonometer as a reliable method and a practical alternative to the Goldmann tonometer. Epstein5 recommended the utilization of the Perkins portable tonometer on obese patients to avoid false IOP elevations due to breath-holding when measured with the slit lamp. This fact is not widely known among practicing ophthalmologists. In an article about sources of error in the use of Goldmann-type tonometers by Whitacre and Stein,19 similar IOP-influencing factors, such as increased venous pressure from tight clothing or from the Valsalva maneuver, were reported, but not the transitory IOP increase in obese patients. We believe simultaneous breath-holding and thorax compression, with subsequent increase in venous pressure, cause transitory elevations of IOP. Because there are great variations in anatomic characteristics that may influence Goldmann tonometry, and IOP is a dynamic phenomenon varying with many factors, we found clinical observation to be the best parameter to check for the patient’s difficulty in positioning at the slit lamp. Clinical signs of the examined patient, such as the compression of the patient against the slit lamp, the effort required to hold the position, breath-holding, and hyperemia of the face due to venous congestion, can easily be watched for during routine IOP measurements. Perkins tonometry in obese pa-
dos Santos et al 䡠 Tonometry in Overweight Patients tients may help avoid a false diagnosis of glaucoma caused by transitorily elevated IOP.
References 1. Moses RA. Fluorescein in applanation tonometry. Am J Ophthalmol 1960;49:1149 –55. 2. Moses RA. The Goldmann applanation tonometer. Am J Ophthalmol 1958;46:865–9. 3. Motolko MA, Feldman F, Hyde M, Hudy D. Sources of variability in the results of applanation tonometry. Can J Ophthalmol 1982;17:93–5. 4. Booth FM, Saad R, Tait SA. Tonography, Valsalva maneuver and central visual fields. A study of 429 patients. Aust N Z J Ophthalmol 1991;19:53–9. 5. Epstein DL. Chandler and Grant’s Glaucoma, 3rd ed. Philadelphia: Lea & Febiger, 1986;14 –31. 6. The Advanced Glaucoma Intervention Study (AGIS): 1. Study design and methods and baseline characteristics of study patients. Control Clin Trials 1994;15:299 –325. 7. Perkins ES. Hand-held applanation tonometer. Br J Ophthalmol 1965;49:591–3. 8. Krieglstein GK, Waller WK. Goldmann applanation versus hand-applanation and Schio¨tz indentation tonometry. Albrecht von Graefes Arch Klin Exp Ophthalmol 1975;194:11– 6.
9. Guyton AC, Jones CE. Central venous pressure: physiological significance and clinical implications. Am Heart J 1973;86: 431–7. 10. Alfano JE, Alfano PA. Glaucoma and the superior vena caval obstruction syndrome. Am J Ophthalmol 1956;42: 685–96. 11. Rosen DA, Johnston VC. Ocular pressure patterns in the Valsalva maneuver. Arch Ophthalmol 1959;62:810 – 6. 12. Bain WES, Maurice DM. Physiological variations in the intraocular pressure. Trans Ophthalmol Soc UK 1959;79:249 – 60. 13. Grehn F, Mackensen G. Die Glaukome, 11th ed. Stuttgart: Kolhammer, 1993;127–38. ¨ ber den spontanen Druckabfall bei Appla14. Bechrakis E. U nationstonometric Ophthalmologica 1966;151:604 –14. 15. Krakau CET, Wilke K. On repeated tonometry. Acta Ophthalmol (Copenh) 1971;49:611– 4. 16. Whitty HPB. Trial results of a hand-held applanation apparatus. Br J Ophthalmol 1969;53:664 –98. 17. Dunn JS, Brubaker RF. Perkins applanation tonometer. Clinical and laboratory evaluation. Arch Ophthalmol 1973;89: 149 –51. 18. Wallace J, Lovell HG. Perkins hand-held applanation tonometer. A clinical evaluation. Br J Ophthalmol 1968;52: 568 –72. 19. Whitacre MM, Stein R. Sources of error with use of Goldmann type-tonometers. Surv Ophthalmol 1993;38:1–30.
2263
2260
epithelial abnormalities, corneal thickness, and astigmatism influence the results of IOP measurements with the Goldmann applanation tonometer.2 Contact between the applanating prism and the eyelashes induces an increase in the apparent IOP. Repeated tonometry induces a decline in IOP.3 Changes in blood circulation, such as increased venous pressure from a Valsalva maneuver, increase IOP.4 Breath-holding may also be associated with the Valsalva maneuver, causing an increase of IOP, and may be an additional source of significant error in applanation tonometry. This phenomenon can be observed, especially in overweight patients who have difficulties in taking up the correct position at the slit lamp during IOP measurements because of a short, stocky build, short neck, and a large chest and abdomen. Epstein5 suggested using the Perkins hand-held tonometer on obese patients to avoid this increase in IOP induced by slit-lamp tonometry. In our glaucoma service, we became aware of a series of obese patients who had persistently elevated IOP for years but no visual field defects and no optic disc pathology. All had been treated for glaucoma for years. Using the Perkins hand-held tonometer,
dos Santos et al 䡠 Tonometry in Overweight Patients the IOP values were within the normal range. In this study, we investigated the extent of the IOP increase caused by anatomic and physiologic factors when using Goldmann applanation tonometry in obese patients.
Patients and Methods We prospectively investigated anatomic and physiologic factors influencing IOP differences between slit-lamp-mounted Goldmann applanation tonometry and Perkins hand-held tonometry in patients with anatomic difficulties in positioning the head at the slit lamp for IOP measurements. Patients were divided into two groups. Group 1 consisted of 12 overweight patients (10 female and 2 male) with positioning difficulties. The mean age was 67.7 ⫾ 8 years (mean ⫾ standard deviation; range, 53–78 years). The patients in this group had a body habitus characterized by a stocky build, a broad face, short neck, and a large chest and abdomen and could hold the head at the slit lamp only with great effort. Nine patients in this group had been treated for primary open-angle glaucoma for 2 to 19 years. Three patients were suspected of having glaucoma. All 12 of these patients had a normal cup– disc ratio and normal visual fields. Three weeks before beginning the study, the patients stopped using all glaucoma medication. The IOP values of these patients were compared to those of subjects of a control group (group 2), which consisted of 70 subjects who had no difficulty in positioning the head at the slit lamp. Forty-five of these subjects had no ocular pathology. Five patients had newly diagnosed glaucoma without any treatment, and 20 patients had primary open-angle glaucoma without filtering surgery. The average age for the control group was 64.9 ⫾ 12 years (range, 33– 89 years). Clinical signs such as the compression of the patient against the slit lamp, effort required to hold the position, breath-holding, and hyperemia of the face due to venous congestion were considered. “Borderline” patients, in which there was doubt as to whether they were having difficulty in positioning at the slit lamp, were excluded from both groups. Body mass index (BMI) was used as a method to define obesity but was not used to divide the participants into the two groups. BMI was calculated with the formula: weight (kg)/height (m2). A BMI between 20 and 25 is considered to be within the normal range. Obesity is taken to start at a BMI of 30. The group of overweight patients had an average BMI of 34.9 ⫾ 3.82, corresponding to an average weight of 91.8 kg and a height of 162 cm. The control group had an average BMI of 24.9 ⫾ 3.45, corresponding to an average weight of 70.1 kg and a height of 169 cm. Intraocular pressure was measured with the Goldmann applanation tonometer on a slit-lamp biomicroscope (BQ 900; Haag– Streit AG, Bern, Switzerland; slit-lamp table: Block, Dortmund, Table 1. Intraocular Pressure Measurements with the Perkins Hand-held Tonometer and the Goldmann Slit-lamp Mounted Tonometer of the Overweight Patients (n ⫽ 12) Goldmann
Mean SD Minimum Maximum Range
Perkins
RE
LE
RE
LE
20.6 2.28 18.0 26.0 8
21.0 3.16 16.0 28.0 12
16.3 2.39 13.0 20.0 7
16.3 2.42 11.0 19.0 8
RE ⫽ right eye; LE ⫽ left eye; SD ⫽ standard deviation.
Table 2. Intraocular Pressure Measurements of the Perkins Hand-held Tonometer and the Goldmann Slit-lamp Mounted Tonometer of the Control Subjects (n ⫽ 70) Goldmann
Mean SD Minimum Maximum Range
Perkins
RE
LE
RE
LE
16.8 3.71 8 26.0 21
18.0 4.2 10.0 32.0 22
16.4 3.72 8.0 26.0 21
17.7 4.33 10.0 32.0 22
RE ⫽ right eye; LE ⫽ left eye; SD ⫽ standard deviation.
Germany) using the method of the Advanced Glaucoma Intervention Study, and with the Perkins hand-held applanation tonometer (Clement Clarke International Ltd, Harlow, England).6,7 Perkins and Goldmann tonometers were calibrated before the beginning of the study. The IOP measurements were performed starting either with Goldmann or with Perkins tonometry and then reversing the sequence after an interval of 10 minutes. All patients measured with the Perkins tonometer were measured in the sitting position. By keeping the test conditions constant, we tried to minimize physiologic changes in IOP other than changes due to the efforts of the overweight patients to position themselves at the slit lamp. No respiratory maneuvers or special movements were made to try to modify or normalize the IOP. The Wilcoxon rank–sum test was used to analyze differences. A P value less than 0.001 was considered statistically significant.
Results In the group of obese patients, the mean IOP was 20.6 ⫾ 2.28 mmHg (range, 18 –26 mmHg) for the right eye and 21.0 ⫾ 3.16 mmHg (range, 16 –28 mmHg) for the left eye as determined by Goldmann tonometry, and 16.3 ⫾ 2.39 mmHg (range, 13–20 mmHg) for the right eye and 16.3 ⫾ 2.42 mmHg (range, 11–19 mmHg) for the left eye when determined by Perkins tonometry. The mean IOP with the Goldmann tonometer of the control group was 16.8 ⫾ 3.71 mmHg (range, 8 –26 mmHg) for the right eye and 18.0 ⫾ 4.2 mmHg (range, 10 –27 mmHg) for the left eye. Very similar IOP values were obtained in this group with the Perkins tonometer with 16.4 ⫾ 3.72 mmHg (range, 8 –26 mmHg) for the right eye and 17.7 ⫾ 4.33 mmHg (range, 10 –27 mmHg) for the left eye (Tables 1 and 2). When analyzing the mean differences between the two tonometers in the obese patients, we found a significant IOP decrease of 4.5 ⫾ 1.3 mmHg (range, 3–7 mmHg) for the right eye and 4.9 ⫾ 1.9 mmHg (range, 2–9 mmHg) for the left eye when measured with the Perkins tonometer. In the control group, the mean difference between the two types of tonometers for the right eye was 0.34 ⫾ 0.69 mmHg and for the left eye was 0.33 ⫾ 0.82 mmHg, which was not statistically significant. When comparing both groups, these values were statistically significant for both eyes (P ⬍ 0.001). In the group of obese patients, the mean BMI of 34.9 ⫾ 3.82 (range, 28.5– 41.9) was significantly lower (P ⬍ 0.001) compared to the control group (24.9 ⫾ 3.45; range, 20.1–33.6). No statistically significant difference was observed regarding the age between the two groups. There was a 5:1 ratio of women to men in the group of obese patients, whereas the control group had an equal gender distribution.
2261
Ophthalmology Volume 105, Number 12, December 1998
Discussion Although slit-lamp-mounted Goldmann applanation tonometry is the most common way to measure IOP and is characterized by good accuracy and reproducibility, ophthalmologists should be aware of sources of underestimation or overestimation to avoid false diagnosis and unnecessary treatment of glaucoma.8 This study describes a series of overweight patients in our glaucoma service who had falsely been diagnosed as having glaucoma. The IOP was significantly higher in these patients when measured with Goldmann tonometry compared to Perkins hand-held tonometry. Anatomic and mechanical difficulties in putting the chin in the right position at the slit lamp cause muscular contraction, breath-holding, and anxiety, with effects on the heart and circulatory system. Pushing thorax and abdomen against the slit lamp and table while breath-holding works like a Valsalva maneuver. Valsalva-like maneuvers frequently are encountered in many daily activities like lifting heavy loads, defecation, playing wind instruments, coughing, and vomiting.9 The Valsalva maneuver, a procedure in which the intrathoracic pressure is raised by an expiratory effort, produces a variety of hemodynamic responses. In patients with a normal cardiac condition, the central venous pressure increases and the systolic and diastolic pressures decrease. Elevation of central venous pressure acts as a back pressure on the systemic circulation and opposes the return of blood from the peripheral blood vessels into the heart. Positive or negative pressures caused by normal or abnormal respiration frequently alter the central venous pressure as much as 4 to 5 mmHg, and occasionally as much as 20 to 30 mmHg. Elevation in venous pressure communicates through the jugular, orbital, and vortex veins to the choroid and can therefore increase the IOP in two different ways: 1. The elevated pressure in the orbital veins interferes with the outflow of aqueous from the anterior chamber. 2. The elevated pressure can cause venous stasis, with vascular engorgement and increase in the volume of the choroidal veins. During the Valsalva maneuver, the acute alteration of the IOP is more likely to be caused by venous stasis with a rise in intraocular volume. Glaucoma in superior vena caval obstruction syndrome is a good clinical example that directly demonstrates the relationship between elevated venous pressure and the IOP.10 The importance of respiratory pressure as a cause of increase in thoracic pressure and subsequent elevation of venous and ocular pressures was shown by Rosen and Johnston,11 who described ocular pressure patterns in the Valsalva maneuver. The force of the expiratory effort was found to have a considerable influence on the magnitude of the ocular pressure response, showing an almost linear increase in IOP with increasing force of expiration. Several patients showed IOP increases up 20 mmHg. Bain and Maurice12 produced an elevation of venous pressure by inflating a sphygmomanometer cuff around the
2262
neck to 40 mmHg, which almost doubled the IOP. Constrictive clothing around the neck may acutely increase the IOP as the neck is extended when positioning the head at the slit lamp.13 Repeated tonometer contact induces a reduction of the IOP.14 –16 The extent of the reduction depends on the number of contacts and on the interval of time between the measurements. To exclude this possible factor of error in our study, IOP measurements were performed in the eyes in a reverse sequence after an interval of 10 minutes. The difference between the decrease of IOP from Goldmann to Perkins tonometry and the rise in the values when the IOP was measured in reverse order was statistically insignificant in both groups. Perkins hand-held tonometry as a method of IOP measurement has been assessed previously.8,17,18 In the study of Whitty,16 all readings taken with the hand applanator were found to be lower than those taken with the Haag–Streit tonometer. A difference of 1.0 mmHg and 1.5 mmHg was thought to be acceptable. In our control group, the mean difference between Perkins and Goldmann tonometry was 0.34 ⫾ 0.69 mmHg for the right eye and 0.33 ⫾ 0.83 mmHg for the left eye, thus confirming good reproducibility. In comparison to the Goldmann tonometer, the Perkins tonometer is more difficult to use. The illumination is weaker, and it is harder to see the fluorescein half-rings compared to the Goldmann tonometer. Despite the relative difficulty for the patients to maintain a steady head posture without a support and for the user to hold the apparatus with a steady hand and wrist, use of the Perkins tonometer can be learned after a short time of practice. The tendency to take slightly lower readings in the case of premature cessation of the tonometry was taken into account. The standard deviation between Goldmann and Perkins tonometers in the control group did not exceed ⫾0.82 mmHg and confirms the Perkins tonometer as a reliable method and a practical alternative to the Goldmann tonometer. Epstein5 recommended the utilization of the Perkins portable tonometer on obese patients to avoid false IOP elevations due to breath-holding when measured with the slit lamp. This fact is not widely known among practicing ophthalmologists. In an article about sources of error in the use of Goldmann-type tonometers by Whitacre and Stein,19 similar IOP-influencing factors, such as increased venous pressure from tight clothing or from the Valsalva maneuver, were reported, but not the transitory IOP increase in obese patients. We believe simultaneous breath-holding and thorax compression, with subsequent increase in venous pressure, cause transitory elevations of IOP. Because there are great variations in anatomic characteristics that may influence Goldmann tonometry, and IOP is a dynamic phenomenon varying with many factors, we found clinical observation to be the best parameter to check for the patient’s difficulty in positioning at the slit lamp. Clinical signs of the examined patient, such as the compression of the patient against the slit lamp, the effort required to hold the position, breath-holding, and hyperemia of the face due to venous congestion, can easily be watched for during routine IOP measurements. Perkins tonometry in obese pa-
dos Santos et al 䡠 Tonometry in Overweight Patients tients may help avoid a false diagnosis of glaucoma caused by transitorily elevated IOP.
References 1. Moses RA. Fluorescein in applanation tonometry. Am J Ophthalmol 1960;49:1149 –55. 2. Moses RA. The Goldmann applanation tonometer. Am J Ophthalmol 1958;46:865–9. 3. Motolko MA, Feldman F, Hyde M, Hudy D. Sources of variability in the results of applanation tonometry. Can J Ophthalmol 1982;17:93–5. 4. Booth FM, Saad R, Tait SA. Tonography, Valsalva maneuver and central visual fields. A study of 429 patients. Aust N Z J Ophthalmol 1991;19:53–9. 5. Epstein DL. Chandler and Grant’s Glaucoma, 3rd ed. Philadelphia: Lea & Febiger, 1986;14 –31. 6. The Advanced Glaucoma Intervention Study (AGIS): 1. Study design and methods and baseline characteristics of study patients. Control Clin Trials 1994;15:299 –325. 7. Perkins ES. Hand-held applanation tonometer. Br J Ophthalmol 1965;49:591–3. 8. Krieglstein GK, Waller WK. Goldmann applanation versus hand-applanation and Schio¨tz indentation tonometry. Albrecht von Graefes Arch Klin Exp Ophthalmol 1975;194:11– 6.
9. Guyton AC, Jones CE. Central venous pressure: physiological significance and clinical implications. Am Heart J 1973;86: 431–7. 10. Alfano JE, Alfano PA. Glaucoma and the superior vena caval obstruction syndrome. Am J Ophthalmol 1956;42: 685–96. 11. Rosen DA, Johnston VC. Ocular pressure patterns in the Valsalva maneuver. Arch Ophthalmol 1959;62:810 – 6. 12. Bain WES, Maurice DM. Physiological variations in the intraocular pressure. Trans Ophthalmol Soc UK 1959;79:249 – 60. 13. Grehn F, Mackensen G. Die Glaukome, 11th ed. Stuttgart: Kolhammer, 1993;127–38. ¨ ber den spontanen Druckabfall bei Appla14. Bechrakis E. U nationstonometric Ophthalmologica 1966;151:604 –14. 15. Krakau CET, Wilke K. On repeated tonometry. Acta Ophthalmol (Copenh) 1971;49:611– 4. 16. Whitty HPB. Trial results of a hand-held applanation apparatus. Br J Ophthalmol 1969;53:664 –98. 17. Dunn JS, Brubaker RF. Perkins applanation tonometer. Clinical and laboratory evaluation. Arch Ophthalmol 1973;89: 149 –51. 18. Wallace J, Lovell HG. Perkins hand-held applanation tonometer. A clinical evaluation. Br J Ophthalmol 1968;52: 568 –72. 19. Whitacre MM, Stein R. Sources of error with use of Goldmann type-tonometers. Surv Ophthalmol 1993;38:1–30.
2263