- Email: [email protected]
Premature Ventricular Systole Detection by Applanation Tonometry
PREMATURE VENTRICULAR SYSTOLE D E T E C T I O N BY A P P L A N A T I O N T O N O M E T R Y P A U L R. L I C H T E R , M . D . , A N D T E R R Y J. B E R G S T R O M , M . D .
Ann Arbor, Michigan Premature ventricular beats can be identified during tonography and they are often found in asymptomatic patients. 1 , 2 These irregular beats result in a decrease in total ocular blood volume which is measured tonographically as a decrease in intraocular pressure. They are manifested in the tonographic tracing as short, sharp, downward deflections with a return to pre-deflection levels during several sub sequent normal heartbeats. 3 · 4 Theoreti cally, similar changes should be recog nized as well with applanation tonometry as with indentation tonography. Yet no mention has been made of identification of these cardiac irregularities during ap planation tonometry either as a potential source of error of applanation tonometry or as a means of screening for possible cardiac disease. We have seen many patients whose irregular heartbeats were demonstrated during Goldmann applanation tonome try. In six of these patients, we initially discovered premature ventricular beats during applanation tonometry. CASE REPORTS
Case 1—A 55-year-old woman was seen in De cember 1971 for evaluation of elevated intraocular pressure. Her medications included only corticosteroid eyedrops for one week for conjunctivitis. She had no known cardiac problems or symptoms and had not had a known previous electrocardiogram (ECG). Intraocular pressures by applanation tonom etry were R.E.: 26 mm Hg and L.E.: 30 mm Hg, but ocular pulsations varied and radial artery palpation demonstrated premature systoles as often as every three heartbeats. Corticosteroid eyedrops were stopped and 1% pilocarpine eyedrops were initiated
From the Department of Ophthalmology, Univer sity Hospital, Ann Arbor, Michigan. Reprint requests to Paul R. Lichter, M.D., Depart ment of Ophthalmology, C6189, University Hospi tal, Ann Arbor, MI 48104.
in each eye four times daily. On follow-up examina tion later that month, intraocular pressures were R.E.: 11 to 17 mm Hg, and L.E.: 17 to 25 mm Hg, the variations occurring with her premature systoles. Tonography also demonstrated many premature sys toles that caused a decrease in intraocular pressure (2 to 7 mm Hg). Three to five normal heartbeats were then needed to return to pre-deflection levels (Fig. 1, a). An ECG performed later that day was interpreted as normal: it showed no arrhythmias (Fig. 1, b). The patient was seen periodically until 1974 with the same asymptomatic findings. She then was lost to follow-up. Case 2—A 52-year-old man was seen in March 1972 for evaluation of his glaucoma therapy. His only medications were 2% pilocarpine eyedrops in both eyes, four times daily, and a daily vitamin C tablet. He had no known cardiac problems or symp toms and had not had a known abnormal ECG in the past. On applanation tonometry, intraocular pres sures were R.E.: 27 mm Hg, and L.E.: 24 mm Hg, with periodic variations in the tonometry end points. This variability coincided with suspected premature ventricular beats on palpation of the radial pulse. Tonography also demonstrated these premature systoles (Fig. 2, a). They were often bigeminal, were associated with decreases of intra ocular pressures (2 to 3 mm Hg), and required three to four normal heartbeats to return to pre-deflection levels. Since there appeared to be an on-off correla tion between initiation of these premature systoles and application of either the tonometry or tono graphy instrument to the eyes, a simultaneous ECG rhythm strip performed before (Fig. 2, b) and during (Fig. 2, c) instrument application to the eye demon strated that the premature systoles were provoked by these procedures. The patient was then directed to use 2% epinephrine hydrochloride eyedrops in both eyes, twice daily. In one week an ECG demonstrat ed a sinus arrhythmia with frequent premature ven tricular beats, often in a bigeminal pattern. The eyedrops were discontinued (L.E. in Novem ber 1972, and R.E. in March 1973) after successful filtering surgery. In 1974 the patient noticed symp toms of increasing fatigue, rapid heartbeat, and shortness of breath. An ECG showed a pattern of atrial fibrillation. The patient has been maintained on a digitalis preparation since then and his symp toms have cleared. When last seen in June 1975, he had frequent extra systoles on applanation tonome try but they were not provoked either by applanation tonometry or by tonography (Fig. 2, d). By applana tion tonometry, intraocular pressures were R.E.: 4 mm Hg, and L.E.: 6 mm Hg, and they decreased about 1 mm Hg with the abnormal systoles. Case 3—A 56-year-old man was seen in March 1972 for evaluation of recurrent uveitis and secon-
797
798
AMERICAN JOURNAL OF OPHTHALMOLOGY
JUNE, 1976
Fig. 1 (Lichter and Bergstrom). Case 1. a, Tonography demonstrates the same premature systoles and decreases in pressure as during the initial applanation tonometry test. Arrows (RE) denote premature systoles. Arrows (LE) show runs of trigeminal (3) rhythm and bigeminal rhythm (2). b, A later ECG was interpreted as normal.
Fig. 2 (Lichter and Bergstrom). Case 2. a, Tonography demonstrates many premature systoles with runs of bigeminal rhythm (arrows), as noted during applanation tonometry. b, Pretonography ECG shows no abnormality, c, An ECG during tonography demonstrates run of bigeminal rhythm, as noted during tonometry. d, Recent ECG with patient receiving treatment for atrial fibrillation. No differences were noted during tonometry or tonography.
VOL. 81, NO. 6
PREMATURE VENTRICULAR
dary glaucoma in the left eye. He was receiving 0.12% prednisolone acetate eyedrops, four times daily and 5% homatropine eyedrops, four times daily, to his left eye, and 250 mg of acetazolamide orally, four times daily. He had no known cardiac problems or symptoms and no cardiac abnormalities were noted on recent physical examinations al though no ECG had been performed. By applanation tonometry, intraocular pressures were R.E.: 19 mm Hg, and L.E.: 23 mm Hg, and there were many premature systoles. Subsequently he was given 1% epinephrine hydrochloride eyedrops, twice daily, to the left eye. In October 1972, he developed symp toms of shortness of breath which were thought to be secondary to the acetazolamide treatment. A decrease in this medication cleared these symptoms. A subsequent internal medicine evaluation and E C G revealed three premature ventricular beats during his E C G testing period. He later developed a local epinephrine reaction and the epinephrine hy drochloride was discontinued in February 1974 without any change in his ECG. In March 1974,4% pilocarpine was started in both eyes but was stopped in the left eye in May. He underwent a cataract extraction in his left eye in January 1975. He has continued to use 4% pilocarpine eyedrops in the right eye, and 0.12% prednisolone acetate and 5% homatropine in the left eye since his cataract extrac tion. An E C G in January 1975 showed .multifocal premature beats. By applanation tonometry, intraoc ular pressures were R.E.: 15 to 17 mm Hg, a n d L . E . : 12 to 14 mm Hg. Tonography showed the decrease in intraocular pressures to be only 1 mm Hg during the premature beats, but three to four normal heart beats were still required (as with applanation to nometry) to return to pre-deflection levels. In this patient, we noticed more premature systo les during the initial phases of applanation tonome-
SYSTOLE
799
try than on later parts. He also had more premature systoles during applanation tonometry on his first visits to our office than he had on the last few visits, although, on routine ECG, the number of premature beats has increased over the years. At the last simul taneous applanation tonometry/ECG test, he dem onstrated more premature contractions while putting drops in his eyes than he did during the tonometry itself. There were more premature ven tricular beats in the operated left eye (whose eyelid had to be held up during tonometry) than in the right eye, which was tested first and needed no eyelid elevation (Fig. 3). Case 4—A 67-year-old man was seen in May 1974 for a glaucoma evaluation. His only medications then were 1% pilocarpine eyedrops once a day in each eye. He had no known heart problems or symptoms and no known previous ECG abnormali ties. Intraocular pressures were R.E.: 23 mm Hg, and L.E.: 28 mm Hg, with noticeable multiple premature systoles. Pilocarpine eyedrops were in creased, four times daily, and he was seen three weeks later. Tonography demonstrated only two premature beats and these were both seen on the tracing from the first eye (right eye). Intraocular pressure decreased by 4 mm Hg, and the tracing returned to former levels in five to six normal heartbeats. We added 2% epinephrine hydrochlo ride eyedrops, twice daily, to the right eye. Later, carbonic anhydrase inhibitors were used and the intraocular pressures were R.E.: 13 mm Hg, and L.E.: 14 mm Hg, in February 1975; but several premature systoles were still noted during the ap planation procedure. An E C G showed sinus bradycardia with occasional premature ventricular beats. Since February 1975, applanation-induced prema ture systoles have been less and less demonstrable, although still present (Fig. 4, a). They typically
Fig. 3 (Lichter and Bergstrom). Case 3. a, Premature beats occurring as anesthetic drops were instilled into the eyes, b, Premature beats occur less often during applanation tonometry on the right eye. c, Premature beats occur more often during tonometry on the left eye when eyelids were held open.
800
AMERICAN JOURNAL OF OPHTHALMOLOGY
^ Q V ^ ^ W ^JV^I^B
JUNE, 1976
*d*~J{-,—4—*.
Fig. 4 (Lichter and Bergstrom), a, Case 4. An occasional premature systole occurred during applanation tonometry. b, Case 5. Runs of bigeminy during tonometry. c, Case 6. Premature systoles during tonometry. d, Case 6. Strip from later ECG showing no premature beats. occurred early during the applanation procedure and then tapered down to their usual (ECG) preva lence. However, intraocular pressures were de creased by 4 to 5 mm Hg, and they returned to former levels after three to five heartbeats. Case 5—A 54-year-old man was referred in May 1974 for evaluation of optic disk abnormalities. His medications included only 0.5% pilocarpine eyedrops, four times daily, and 1% epinephrine hydrochloride eyedrops, twice daily, to both eyes, and 1 grain of thyroid orally daily. He had no known cardiac problems or symptoms and had no knowl edge of any abnormality on any prior ECG. Intrao cular pressures by applanation tonometry were R.E.: 13 mm Hg, and L.E.: 14 mm Hg, and no irregular pulsations were observed. Tonography was per formed and showed no premature beats. He was subsequently admitted to the hospital in June 1974 for surgical excision of a pituitary adenoma. On admission, an ECG was normal. In September 1974, he was receiving 30 mg of hydrocortisone daily, 2 grains of thyroid daily, and 100 mg of diphenylhydantoin, three times daily. There were no premature systoles on applanation tonometry. He was seen again in April 1975 for evaluation of visual field loss secondary to his neurosurgical procedure. He was taking the same medications, and intraocular pres sures were R.E.: 15 to 19 mm Hg, and L.E.: 16 to 19 mm Hg, with many premature systoles. By the time he was placed on the ECG monitor for simultaneous
tonometry, tonography, and ECG, he was in a bigeminal rhythm pattern that persisted for the re mainder of the examination (Fig. 4, b). By tono graphy, the pressure decreased by about 2 mm Hg. An ECG performed by the patient's local internist was normal in June 1975, and he has not demon strated premature beats during any other examina tions. Case 6—A 58-year-old man was referrred to us in June 1975 for evaluation of one episode of anterior uveitis associated with an increased intraocular pressure in his left eye. He had been treated success fully and, at the time of our evaluation, was asymp tomatic and receiving no medications. He had no known heart problems or symptoms and knew of no abnormalities on any previous ECG. By applanation tonometry, intraocular pressures were R.E.: 20 mm Hg, and L.E.: 15 mm Hg, and many premature systoles were noted on initial tonometry, as often as every two to four beats in either eye. Intraocular pressures decreased by 3 to 4 mm Hg each time, and three to five subsequent normal heartbeats were required to recover. By the time the ECG rhythm strip could be attached, the number of premature beats during tonometry dropped rapidly (Fig. 4, c) and during subsequent tonography, only one prema ture systole was noted and that one was near the end of the second tracing (after eight minutes of testing) and resulted in a decrease of only 1 mm Hg. An ECG later in the day was normal and did not
VOL. 81, NO. 6
PREMATURE VENTRICULAR SYSTOLE
demonstrate any premature beats (Fig. 4, d). We have not seen the patient since his initial visit. DISCUSSION
Applanation tonometry has been wide ly used in ophthalmology since it was described by Goldmann 5 in 1955. To measure the intraocular pressure by Goldmann's method, a prism is used to split a circle into two halves. The instru ment contacts the cornea and, enhanced by fluorescein, the tears are seen as two halves of a circle. When the inner aspect of the left side of the lower semicircle meets the inner edge of the right side of the upper simicircle, a reading is taken as the tonometry value (Fig. 5, a). However, in all dynamic tests involving intraocular pressure, with each systole and diastole, changes occur in ocular blood volume that create an ocular pulse. This pulse can be seen on a tonographic recording as regular up and down movements of the graph. Intraocular pressure increases and decreases with each systole and diastole. The ocular pulse is also observed dur ing applanation tonometry as the mires move back and forth with each pulse. In patients with a wide ocular pulse, there
801
can be confusion as to the end point of tonometry. However, in some patients the examiner observes a regular pulsation of the mires that suddenly changes. The semicircles enlarge and separate to indi cate that the intraocular pressure has de creased (Fig. 5, b). Then, gradually, as the pulsations are observed, the mires again assume their previous end-point position. This sudden decrease in pressure is likely caused by a premature ventricular beat. Then, as the cardiac output recovers, the intraocular pressure returns to its former level, only to decrease again with repeat premature ventricular beats. Cardiologists consider premature ven tricular beats as common findings even in routine electrocardiographic examina tion. 6 - 8 In young people without organic heart disease these beats are often consid ered innocuous. They are, of course, more common in almost any kind of heart dis ease but especially in myocardial infarc tion and congestive heart failure. 9 They are also frequently associated with digi talis intoxication and can occur with hypokalemia, anoxia, and excessive use of tobacco, caffeine, and alcohol. Other causes include emotional and physical
Fig. 5 (Lichter and Bergstrom). Left, Position of mires by Goldmann applanation tonometry at end-point reading of intraocular pressure. Right, Position of mires, by applanation tonometer, and premature ventricular beats showing sudden separation of the mires, indicating a decrease in intraocular pressure.
802
AMERICAN JOURNAL OF OPHTHALMOLOGY
stresses and the use of sympathomimetic drugs—amphetamines, epinephrine, and isoproterenol (even in the forms of nose drops, sprays, and inhalers). 9 Additional systoles occur more often in patients receiving epinephrine for glau coma than in patients who are not receiv ing epinephrine. 3 However, none of our six patients were receiving epinephrine drops during the initial findings. Three patients (Cases 2-4) were subsequently treated with epinephrine drops. One pa tient (Case 2) used them in both eyes for a year and then discontinued them; we noted no significant cardiac symptoms. A year after discontinuing the drops, he developed symptoms of atrial fibrillation and required treatment. Another patient (Case 3) used epinephrine drops in one eye only for almost two years. One year after he discontinued the drops, an E C G demonstrated an increased number of premature systoles rather than fewer sys toles. In one patient (Case 4), the drops were continued in one eye only but no changes in the arrhythmia have been de tected by ECG. Two patients (Cases 1 and 3) were taking corticosteroid drops (Case 1, regu lar strength in both eyes, and Case 3, 0.12% prednisolone acetate in the left eye only); two patients (Cases 2 and 4) were receiving pilocarpine eyedrops (Case 2, 2% in both eyes, four times daily, and Case 4, 1% in both eyes once daily); one patient (Case 3) was also taking 5% homatropine eyedrops in the left eye, and acetazolamide tablets; one patient (Case 5) was taking replacement thyroid and hy drocortisone as well as diphenylhydantoin; and one patient (Case 6) was receiv ing no medications. There seems to be no significant corre lation of drugs with this small series of cardiac arrhythmias, although there was a previously demonstrated correlation of increased numbers of extra systoles in patients receiving epinephrine eyedrops. In five of our six patients (all but Case
JUNE, 1976
2), the decrease in intraocular pressure we observed during applanation tonometry was generally greater than that measured during tonography, but it varied from patient to patient and from heartbeat to heartbeat. Nevertheless, with both meth ods, three to six normal heartbeats were needed to return to pre-deflection levels. There appeared to be little correlation between the amount of the decrease in intraocular pressure and the number of normal heartbeats to return to prior levels. Nor did the amount of decrease seem to be related to the height of the pressure, except in one patient (Case 2) who showed only a slight decrease associated with low intraocular pressures in his postoperative evaluation. In three patients (Cases 3, 4, and 6) there were fewer premature systoles as the ocular examination progressed; dur ing tonography, there were fewer than during the prior tonometry. In one patient (Case 6), there were fewer systoles with tonometry after the E C G monitor was applied than during the initial tonometry; by the time tonography was performed, only one premature beat was noted. In two patients (Cases 3 and 4), there were fewer systoles as the number of return visits increased, although systoles still occurred with greater frequency during the initial applanation tonometry than during an E C G performed later. In three patients (Cases 1, 5, and 6), the subse quent ECGs were normal. In Case 2, there were initially fewer premature beats on the ECG than there were on the rhythm strips accompanying tonometry and tonography. Thus, all six patients appeared to be stimulated by the proce dure of applanation tonometry to produce premature cardiac beats. Three patients had premature ventricular beats only dur ing ocular examination and the other three had increased premature ventricular beats superimposed on a permanent car diac arrhythmia. These findings agree with the well known fact that premature
VOL. 81, NO. 6
PREMATURE VENTRICULAR SYSTOLE
systoles may occur more often during periods of stress, such as an initial applanation tonometry test. The examiner should note carefully the applanation pat terns on the initial pressure checks for any cardiac abnormalities. The significance of this small series of patients is twofold. First, the examiner must be aware that by using applanation tonometry, the intraocular pressure may be 8 mm Hg lower than normal (usually only 2 to 5 mm lower). This could result in a false reassurance of glaucoma control, in an unnecessary investigation of a greater than 4 mm Hg pressure difference between the two eyes, or in not suspect ing glaucoma. Secondly, although the ar rhythmias initially found on applanation tonometry are often considered innocu ous in younger people, in some age groups (especially middle-aged and eld erly persons), they are associated with an increased prevalence of heart disease and sudden death. 8 - 1 0 Since patients with ar rhythmias are often asymptomatic at any age, and since these arrhythmias can be demonstrated on applanation tonometry, ophthalmologists can identify these pa tients early and direct them to a cardiolo gist. Ophthalmologists must be aware that arrhythmias may be present on initial applanation tonometry; the mires should be observed during several pulses. Wide swings in the ocular pulse should be noted and investigated. We have observed premature systoles in many other patients during applana tion tonometry. Since the incidence of premature ventricular beats in the general population may be as high as 3.6%, 8 find ing premature ventricular beats during applanation tonometry should occur often. Since these beats may be induced during this examination, the ophthalmol ogist should watch the mires closely dur ing applanation tonometry, since prema ture ventricular beats may not be found by routine radial pulse palpation.
803
SUMMARY
We first identified (and probably stimu lated) premature ventricular systoles in six patients while performing Goldmann applanation tonometry. One patient de veloped atrial fibrillation and one patient had a follow-up electrocardiogram show ing increasing numbers of these extra systoles without any other abnormalities. Just as in tonography, these were detected during applanation tonometry as rapid decreases in intraocular pressures fol lowed by a gradual return to prior levels. The decrease in measured intraocular pressure was occasionally 8 mm Hg but usually was between 2 and 5 mm Hg. These systoles probably occurred during the first part of applanation tonometry and during the first few office visits. REFERENCES 1. Garner, L. L.: Tonography and the Glaucomas. Springfield, Charles C Thomas, 1965, p p . 123, 138, and 144. 2. Drews, R. C : Manual of Tonography. St. Louis, C. V. Mosby, 1971, p p . 15, 19, 76, and 101. 3. Ballin, N., Becker, B., and Goldman, M. L.: Systemic effects of epinephrine applied topically to the eye. Invest. Ophthalmol. 5:125, 1966. 4. Haik. G. M., Perez, L. F., Reitman, H. S., and Massey, J. Y.: Tonographic tracings in patients with cardiac rhythm disturbances. Am. J. Ophthalmol. 70:929, 1970. 5. Goldmann, H.: Un nouveau tonomètre é aplanation. Bull. Soc. Fr. Ophtalmol. 67:474, 1955. 6. Hiss, R. G., Averill, K. H., and Lamb, L. E.: Electrocardiographic findings in 67,375 asympto matic subjects. 1. Incidence of abnormalities. 3. Ventricular arrhythmia. Am. J. Cardiol. 6:76 and 96, 1960. 7. Hiss, R. G., and Lamb, L. E.: Electrocardio graphic findings in 122,043 individuals. Circulation 25:947, 1962. 8. Chiang, B. N., Perlman, L. V., Ostrander, L. D., and Epstein, F. H.: Relationship of premature systo les to coronary heart disease and sudden death in the Tecumseh epidemiologic study. Ann. Intern. Med. 70:1159, 1969. 9. Hurst, J. W., and Logue, B. R.: Distrubances of Cardiac Rhythm and Conduction; Syncope; Shock; and Sudden Death in the Heart. New York, McGraw-Hill Book Co., Inc., 1970, p p . 500 and 552. 10. Pell, S., and D'Alonzo, C. A.: Immediate mor tality and five year survival of employed men with first myocardial infarction. N. Engl. J. Med. 270:915, 1964.
Ann Arbor, Michigan Premature ventricular beats can be identified during tonography and they are often found in asymptomatic patients. 1 , 2 These irregular beats result in a decrease in total ocular blood volume which is measured tonographically as a decrease in intraocular pressure. They are manifested in the tonographic tracing as short, sharp, downward deflections with a return to pre-deflection levels during several sub sequent normal heartbeats. 3 · 4 Theoreti cally, similar changes should be recog nized as well with applanation tonometry as with indentation tonography. Yet no mention has been made of identification of these cardiac irregularities during ap planation tonometry either as a potential source of error of applanation tonometry or as a means of screening for possible cardiac disease. We have seen many patients whose irregular heartbeats were demonstrated during Goldmann applanation tonome try. In six of these patients, we initially discovered premature ventricular beats during applanation tonometry. CASE REPORTS
Case 1—A 55-year-old woman was seen in De cember 1971 for evaluation of elevated intraocular pressure. Her medications included only corticosteroid eyedrops for one week for conjunctivitis. She had no known cardiac problems or symptoms and had not had a known previous electrocardiogram (ECG). Intraocular pressures by applanation tonom etry were R.E.: 26 mm Hg and L.E.: 30 mm Hg, but ocular pulsations varied and radial artery palpation demonstrated premature systoles as often as every three heartbeats. Corticosteroid eyedrops were stopped and 1% pilocarpine eyedrops were initiated
From the Department of Ophthalmology, Univer sity Hospital, Ann Arbor, Michigan. Reprint requests to Paul R. Lichter, M.D., Depart ment of Ophthalmology, C6189, University Hospi tal, Ann Arbor, MI 48104.
in each eye four times daily. On follow-up examina tion later that month, intraocular pressures were R.E.: 11 to 17 mm Hg, and L.E.: 17 to 25 mm Hg, the variations occurring with her premature systoles. Tonography also demonstrated many premature sys toles that caused a decrease in intraocular pressure (2 to 7 mm Hg). Three to five normal heartbeats were then needed to return to pre-deflection levels (Fig. 1, a). An ECG performed later that day was interpreted as normal: it showed no arrhythmias (Fig. 1, b). The patient was seen periodically until 1974 with the same asymptomatic findings. She then was lost to follow-up. Case 2—A 52-year-old man was seen in March 1972 for evaluation of his glaucoma therapy. His only medications were 2% pilocarpine eyedrops in both eyes, four times daily, and a daily vitamin C tablet. He had no known cardiac problems or symp toms and had not had a known abnormal ECG in the past. On applanation tonometry, intraocular pres sures were R.E.: 27 mm Hg, and L.E.: 24 mm Hg, with periodic variations in the tonometry end points. This variability coincided with suspected premature ventricular beats on palpation of the radial pulse. Tonography also demonstrated these premature systoles (Fig. 2, a). They were often bigeminal, were associated with decreases of intra ocular pressures (2 to 3 mm Hg), and required three to four normal heartbeats to return to pre-deflection levels. Since there appeared to be an on-off correla tion between initiation of these premature systoles and application of either the tonometry or tono graphy instrument to the eyes, a simultaneous ECG rhythm strip performed before (Fig. 2, b) and during (Fig. 2, c) instrument application to the eye demon strated that the premature systoles were provoked by these procedures. The patient was then directed to use 2% epinephrine hydrochloride eyedrops in both eyes, twice daily. In one week an ECG demonstrat ed a sinus arrhythmia with frequent premature ven tricular beats, often in a bigeminal pattern. The eyedrops were discontinued (L.E. in Novem ber 1972, and R.E. in March 1973) after successful filtering surgery. In 1974 the patient noticed symp toms of increasing fatigue, rapid heartbeat, and shortness of breath. An ECG showed a pattern of atrial fibrillation. The patient has been maintained on a digitalis preparation since then and his symp toms have cleared. When last seen in June 1975, he had frequent extra systoles on applanation tonome try but they were not provoked either by applanation tonometry or by tonography (Fig. 2, d). By applana tion tonometry, intraocular pressures were R.E.: 4 mm Hg, and L.E.: 6 mm Hg, and they decreased about 1 mm Hg with the abnormal systoles. Case 3—A 56-year-old man was seen in March 1972 for evaluation of recurrent uveitis and secon-
797
798
AMERICAN JOURNAL OF OPHTHALMOLOGY
JUNE, 1976
Fig. 1 (Lichter and Bergstrom). Case 1. a, Tonography demonstrates the same premature systoles and decreases in pressure as during the initial applanation tonometry test. Arrows (RE) denote premature systoles. Arrows (LE) show runs of trigeminal (3) rhythm and bigeminal rhythm (2). b, A later ECG was interpreted as normal.
Fig. 2 (Lichter and Bergstrom). Case 2. a, Tonography demonstrates many premature systoles with runs of bigeminal rhythm (arrows), as noted during applanation tonometry. b, Pretonography ECG shows no abnormality, c, An ECG during tonography demonstrates run of bigeminal rhythm, as noted during tonometry. d, Recent ECG with patient receiving treatment for atrial fibrillation. No differences were noted during tonometry or tonography.
VOL. 81, NO. 6
PREMATURE VENTRICULAR
dary glaucoma in the left eye. He was receiving 0.12% prednisolone acetate eyedrops, four times daily and 5% homatropine eyedrops, four times daily, to his left eye, and 250 mg of acetazolamide orally, four times daily. He had no known cardiac problems or symptoms and no cardiac abnormalities were noted on recent physical examinations al though no ECG had been performed. By applanation tonometry, intraocular pressures were R.E.: 19 mm Hg, and L.E.: 23 mm Hg, and there were many premature systoles. Subsequently he was given 1% epinephrine hydrochloride eyedrops, twice daily, to the left eye. In October 1972, he developed symp toms of shortness of breath which were thought to be secondary to the acetazolamide treatment. A decrease in this medication cleared these symptoms. A subsequent internal medicine evaluation and E C G revealed three premature ventricular beats during his E C G testing period. He later developed a local epinephrine reaction and the epinephrine hy drochloride was discontinued in February 1974 without any change in his ECG. In March 1974,4% pilocarpine was started in both eyes but was stopped in the left eye in May. He underwent a cataract extraction in his left eye in January 1975. He has continued to use 4% pilocarpine eyedrops in the right eye, and 0.12% prednisolone acetate and 5% homatropine in the left eye since his cataract extrac tion. An E C G in January 1975 showed .multifocal premature beats. By applanation tonometry, intraoc ular pressures were R.E.: 15 to 17 mm Hg, a n d L . E . : 12 to 14 mm Hg. Tonography showed the decrease in intraocular pressures to be only 1 mm Hg during the premature beats, but three to four normal heart beats were still required (as with applanation to nometry) to return to pre-deflection levels. In this patient, we noticed more premature systo les during the initial phases of applanation tonome-
SYSTOLE
799
try than on later parts. He also had more premature systoles during applanation tonometry on his first visits to our office than he had on the last few visits, although, on routine ECG, the number of premature beats has increased over the years. At the last simul taneous applanation tonometry/ECG test, he dem onstrated more premature contractions while putting drops in his eyes than he did during the tonometry itself. There were more premature ven tricular beats in the operated left eye (whose eyelid had to be held up during tonometry) than in the right eye, which was tested first and needed no eyelid elevation (Fig. 3). Case 4—A 67-year-old man was seen in May 1974 for a glaucoma evaluation. His only medications then were 1% pilocarpine eyedrops once a day in each eye. He had no known heart problems or symptoms and no known previous ECG abnormali ties. Intraocular pressures were R.E.: 23 mm Hg, and L.E.: 28 mm Hg, with noticeable multiple premature systoles. Pilocarpine eyedrops were in creased, four times daily, and he was seen three weeks later. Tonography demonstrated only two premature beats and these were both seen on the tracing from the first eye (right eye). Intraocular pressure decreased by 4 mm Hg, and the tracing returned to former levels in five to six normal heartbeats. We added 2% epinephrine hydrochlo ride eyedrops, twice daily, to the right eye. Later, carbonic anhydrase inhibitors were used and the intraocular pressures were R.E.: 13 mm Hg, and L.E.: 14 mm Hg, in February 1975; but several premature systoles were still noted during the ap planation procedure. An E C G showed sinus bradycardia with occasional premature ventricular beats. Since February 1975, applanation-induced prema ture systoles have been less and less demonstrable, although still present (Fig. 4, a). They typically
Fig. 3 (Lichter and Bergstrom). Case 3. a, Premature beats occurring as anesthetic drops were instilled into the eyes, b, Premature beats occur less often during applanation tonometry on the right eye. c, Premature beats occur more often during tonometry on the left eye when eyelids were held open.
800
AMERICAN JOURNAL OF OPHTHALMOLOGY
^ Q V ^ ^ W ^JV^I^B
JUNE, 1976
*d*~J{-,—4—*.
Fig. 4 (Lichter and Bergstrom), a, Case 4. An occasional premature systole occurred during applanation tonometry. b, Case 5. Runs of bigeminy during tonometry. c, Case 6. Premature systoles during tonometry. d, Case 6. Strip from later ECG showing no premature beats. occurred early during the applanation procedure and then tapered down to their usual (ECG) preva lence. However, intraocular pressures were de creased by 4 to 5 mm Hg, and they returned to former levels after three to five heartbeats. Case 5—A 54-year-old man was referred in May 1974 for evaluation of optic disk abnormalities. His medications included only 0.5% pilocarpine eyedrops, four times daily, and 1% epinephrine hydrochloride eyedrops, twice daily, to both eyes, and 1 grain of thyroid orally daily. He had no known cardiac problems or symptoms and had no knowl edge of any abnormality on any prior ECG. Intrao cular pressures by applanation tonometry were R.E.: 13 mm Hg, and L.E.: 14 mm Hg, and no irregular pulsations were observed. Tonography was per formed and showed no premature beats. He was subsequently admitted to the hospital in June 1974 for surgical excision of a pituitary adenoma. On admission, an ECG was normal. In September 1974, he was receiving 30 mg of hydrocortisone daily, 2 grains of thyroid daily, and 100 mg of diphenylhydantoin, three times daily. There were no premature systoles on applanation tonometry. He was seen again in April 1975 for evaluation of visual field loss secondary to his neurosurgical procedure. He was taking the same medications, and intraocular pres sures were R.E.: 15 to 19 mm Hg, and L.E.: 16 to 19 mm Hg, with many premature systoles. By the time he was placed on the ECG monitor for simultaneous
tonometry, tonography, and ECG, he was in a bigeminal rhythm pattern that persisted for the re mainder of the examination (Fig. 4, b). By tono graphy, the pressure decreased by about 2 mm Hg. An ECG performed by the patient's local internist was normal in June 1975, and he has not demon strated premature beats during any other examina tions. Case 6—A 58-year-old man was referrred to us in June 1975 for evaluation of one episode of anterior uveitis associated with an increased intraocular pressure in his left eye. He had been treated success fully and, at the time of our evaluation, was asymp tomatic and receiving no medications. He had no known heart problems or symptoms and knew of no abnormalities on any previous ECG. By applanation tonometry, intraocular pressures were R.E.: 20 mm Hg, and L.E.: 15 mm Hg, and many premature systoles were noted on initial tonometry, as often as every two to four beats in either eye. Intraocular pressures decreased by 3 to 4 mm Hg each time, and three to five subsequent normal heartbeats were required to recover. By the time the ECG rhythm strip could be attached, the number of premature beats during tonometry dropped rapidly (Fig. 4, c) and during subsequent tonography, only one prema ture systole was noted and that one was near the end of the second tracing (after eight minutes of testing) and resulted in a decrease of only 1 mm Hg. An ECG later in the day was normal and did not
VOL. 81, NO. 6
PREMATURE VENTRICULAR SYSTOLE
demonstrate any premature beats (Fig. 4, d). We have not seen the patient since his initial visit. DISCUSSION
Applanation tonometry has been wide ly used in ophthalmology since it was described by Goldmann 5 in 1955. To measure the intraocular pressure by Goldmann's method, a prism is used to split a circle into two halves. The instru ment contacts the cornea and, enhanced by fluorescein, the tears are seen as two halves of a circle. When the inner aspect of the left side of the lower semicircle meets the inner edge of the right side of the upper simicircle, a reading is taken as the tonometry value (Fig. 5, a). However, in all dynamic tests involving intraocular pressure, with each systole and diastole, changes occur in ocular blood volume that create an ocular pulse. This pulse can be seen on a tonographic recording as regular up and down movements of the graph. Intraocular pressure increases and decreases with each systole and diastole. The ocular pulse is also observed dur ing applanation tonometry as the mires move back and forth with each pulse. In patients with a wide ocular pulse, there
801
can be confusion as to the end point of tonometry. However, in some patients the examiner observes a regular pulsation of the mires that suddenly changes. The semicircles enlarge and separate to indi cate that the intraocular pressure has de creased (Fig. 5, b). Then, gradually, as the pulsations are observed, the mires again assume their previous end-point position. This sudden decrease in pressure is likely caused by a premature ventricular beat. Then, as the cardiac output recovers, the intraocular pressure returns to its former level, only to decrease again with repeat premature ventricular beats. Cardiologists consider premature ven tricular beats as common findings even in routine electrocardiographic examina tion. 6 - 8 In young people without organic heart disease these beats are often consid ered innocuous. They are, of course, more common in almost any kind of heart dis ease but especially in myocardial infarc tion and congestive heart failure. 9 They are also frequently associated with digi talis intoxication and can occur with hypokalemia, anoxia, and excessive use of tobacco, caffeine, and alcohol. Other causes include emotional and physical
Fig. 5 (Lichter and Bergstrom). Left, Position of mires by Goldmann applanation tonometry at end-point reading of intraocular pressure. Right, Position of mires, by applanation tonometer, and premature ventricular beats showing sudden separation of the mires, indicating a decrease in intraocular pressure.
802
AMERICAN JOURNAL OF OPHTHALMOLOGY
stresses and the use of sympathomimetic drugs—amphetamines, epinephrine, and isoproterenol (even in the forms of nose drops, sprays, and inhalers). 9 Additional systoles occur more often in patients receiving epinephrine for glau coma than in patients who are not receiv ing epinephrine. 3 However, none of our six patients were receiving epinephrine drops during the initial findings. Three patients (Cases 2-4) were subsequently treated with epinephrine drops. One pa tient (Case 2) used them in both eyes for a year and then discontinued them; we noted no significant cardiac symptoms. A year after discontinuing the drops, he developed symptoms of atrial fibrillation and required treatment. Another patient (Case 3) used epinephrine drops in one eye only for almost two years. One year after he discontinued the drops, an E C G demonstrated an increased number of premature systoles rather than fewer sys toles. In one patient (Case 4), the drops were continued in one eye only but no changes in the arrhythmia have been de tected by ECG. Two patients (Cases 1 and 3) were taking corticosteroid drops (Case 1, regu lar strength in both eyes, and Case 3, 0.12% prednisolone acetate in the left eye only); two patients (Cases 2 and 4) were receiving pilocarpine eyedrops (Case 2, 2% in both eyes, four times daily, and Case 4, 1% in both eyes once daily); one patient (Case 3) was also taking 5% homatropine eyedrops in the left eye, and acetazolamide tablets; one patient (Case 5) was taking replacement thyroid and hy drocortisone as well as diphenylhydantoin; and one patient (Case 6) was receiv ing no medications. There seems to be no significant corre lation of drugs with this small series of cardiac arrhythmias, although there was a previously demonstrated correlation of increased numbers of extra systoles in patients receiving epinephrine eyedrops. In five of our six patients (all but Case
JUNE, 1976
2), the decrease in intraocular pressure we observed during applanation tonometry was generally greater than that measured during tonography, but it varied from patient to patient and from heartbeat to heartbeat. Nevertheless, with both meth ods, three to six normal heartbeats were needed to return to pre-deflection levels. There appeared to be little correlation between the amount of the decrease in intraocular pressure and the number of normal heartbeats to return to prior levels. Nor did the amount of decrease seem to be related to the height of the pressure, except in one patient (Case 2) who showed only a slight decrease associated with low intraocular pressures in his postoperative evaluation. In three patients (Cases 3, 4, and 6) there were fewer premature systoles as the ocular examination progressed; dur ing tonography, there were fewer than during the prior tonometry. In one patient (Case 6), there were fewer systoles with tonometry after the E C G monitor was applied than during the initial tonometry; by the time tonography was performed, only one premature beat was noted. In two patients (Cases 3 and 4), there were fewer systoles as the number of return visits increased, although systoles still occurred with greater frequency during the initial applanation tonometry than during an E C G performed later. In three patients (Cases 1, 5, and 6), the subse quent ECGs were normal. In Case 2, there were initially fewer premature beats on the ECG than there were on the rhythm strips accompanying tonometry and tonography. Thus, all six patients appeared to be stimulated by the proce dure of applanation tonometry to produce premature cardiac beats. Three patients had premature ventricular beats only dur ing ocular examination and the other three had increased premature ventricular beats superimposed on a permanent car diac arrhythmia. These findings agree with the well known fact that premature
VOL. 81, NO. 6
PREMATURE VENTRICULAR SYSTOLE
systoles may occur more often during periods of stress, such as an initial applanation tonometry test. The examiner should note carefully the applanation pat terns on the initial pressure checks for any cardiac abnormalities. The significance of this small series of patients is twofold. First, the examiner must be aware that by using applanation tonometry, the intraocular pressure may be 8 mm Hg lower than normal (usually only 2 to 5 mm lower). This could result in a false reassurance of glaucoma control, in an unnecessary investigation of a greater than 4 mm Hg pressure difference between the two eyes, or in not suspect ing glaucoma. Secondly, although the ar rhythmias initially found on applanation tonometry are often considered innocu ous in younger people, in some age groups (especially middle-aged and eld erly persons), they are associated with an increased prevalence of heart disease and sudden death. 8 - 1 0 Since patients with ar rhythmias are often asymptomatic at any age, and since these arrhythmias can be demonstrated on applanation tonometry, ophthalmologists can identify these pa tients early and direct them to a cardiolo gist. Ophthalmologists must be aware that arrhythmias may be present on initial applanation tonometry; the mires should be observed during several pulses. Wide swings in the ocular pulse should be noted and investigated. We have observed premature systoles in many other patients during applana tion tonometry. Since the incidence of premature ventricular beats in the general population may be as high as 3.6%, 8 find ing premature ventricular beats during applanation tonometry should occur often. Since these beats may be induced during this examination, the ophthalmol ogist should watch the mires closely dur ing applanation tonometry, since prema ture ventricular beats may not be found by routine radial pulse palpation.
803
SUMMARY
We first identified (and probably stimu lated) premature ventricular systoles in six patients while performing Goldmann applanation tonometry. One patient de veloped atrial fibrillation and one patient had a follow-up electrocardiogram show ing increasing numbers of these extra systoles without any other abnormalities. Just as in tonography, these were detected during applanation tonometry as rapid decreases in intraocular pressures fol lowed by a gradual return to prior levels. The decrease in measured intraocular pressure was occasionally 8 mm Hg but usually was between 2 and 5 mm Hg. These systoles probably occurred during the first part of applanation tonometry and during the first few office visits. REFERENCES 1. Garner, L. L.: Tonography and the Glaucomas. Springfield, Charles C Thomas, 1965, p p . 123, 138, and 144. 2. Drews, R. C : Manual of Tonography. St. Louis, C. V. Mosby, 1971, p p . 15, 19, 76, and 101. 3. Ballin, N., Becker, B., and Goldman, M. L.: Systemic effects of epinephrine applied topically to the eye. Invest. Ophthalmol. 5:125, 1966. 4. Haik. G. M., Perez, L. F., Reitman, H. S., and Massey, J. Y.: Tonographic tracings in patients with cardiac rhythm disturbances. Am. J. Ophthalmol. 70:929, 1970. 5. Goldmann, H.: Un nouveau tonomètre é aplanation. Bull. Soc. Fr. Ophtalmol. 67:474, 1955. 6. Hiss, R. G., Averill, K. H., and Lamb, L. E.: Electrocardiographic findings in 67,375 asympto matic subjects. 1. Incidence of abnormalities. 3. Ventricular arrhythmia. Am. J. Cardiol. 6:76 and 96, 1960. 7. Hiss, R. G., and Lamb, L. E.: Electrocardio graphic findings in 122,043 individuals. Circulation 25:947, 1962. 8. Chiang, B. N., Perlman, L. V., Ostrander, L. D., and Epstein, F. H.: Relationship of premature systo les to coronary heart disease and sudden death in the Tecumseh epidemiologic study. Ann. Intern. Med. 70:1159, 1969. 9. Hurst, J. W., and Logue, B. R.: Distrubances of Cardiac Rhythm and Conduction; Syncope; Shock; and Sudden Death in the Heart. New York, McGraw-Hill Book Co., Inc., 1970, p p . 500 and 552. 10. Pell, S., and D'Alonzo, C. A.: Immediate mor tality and five year survival of employed men with first myocardial infarction. N. Engl. J. Med. 270:915, 1964.